Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
WEB PERSONALIZATION AND RECOMMENDATION MODEL FOR TRUST IN E-COMMERCE WEBSITES
FROM AN INDIAN PERSPECTIVE
A Thesis submitted to SHOBHIT UNIVERSITY, MEERUT For the award of the degree of
DOCTOR OF PHILOSOPHY In
COMPUTER ENGINEERING
By Devendera Agarwal
Under the Supervision of Prof. (Dr.) S.P. Tripathi
& Prof. (Dr.) R.P. Agarwal
Faculty of
Computer Engineering & Information Technology Shobhit University, Meerut-250110
2013
DECLARATION
I, Devendera Agarwal hereby, declare that the work
presented in this thesis entitled Web Personalization and
Recommendation Model for Trust in E-commerce
Websites from an Indian Perspective in fulfillment of the
requirements for the award of degree of Doctor of Philosophy,
submitted in the school of Computer Engineering &
Information Technology at Shobhit University, Modipuram,
Meerut is an authentic record of my own research work
under the supervision of Prof. (Dr.) S.P. Tripathi and
Prof. (Dr.) R.P. Agarwal.
I also declare that the work embodied in the present thesis
(i) is my original work and has not been copied from any
Journal/ thesis/ book, and
(ii) has not been submitted by me for any other
Degree/Diploma.
(Devendera Agarwal)
CERTIFICATE I, Devendera Agarwal hereby, declare that the work
presented in this thesis, entitled “Web Personalization and
Recommendation Model for Trust in E-commerce
Websites from an Indian Perspective” in fulfillment of the
requirements for the award of Degree of Doctor of Philosophy,
submitted in the School of Computer Engineering &
Information Technology at Shobhit University, Modipuram,
Meerut is an authentic record of my own research work
carried out.
Signature of Candidate
This is to certify that the details given by the research scholar
are true to best of our knowledge.
Prof. (Dr.) S.P. Tripathi Prof. (Dr.) R.P. Agarwal
(Supervisor) (Administrative Supervisor)
CERTIFICATE This is to certify that the thesis entitled “Web
Personalization and Recommendation Model for Trust in
E-commerce Websites from an Indian Perspective”
submitted by Devendera Agarwal for the award of degree of
Doctor of Philosophy in the School of Computer Engineering
& Information Technology of Shobhit University, Meerut is
a record of authentic work carried out by him under our
supervision.
To the best of my knowledge, the matter embodied in this
thesis is the original work of the candidate and has not been
submitted for the award of any other degree or diploma.
It is further certified that he has worked with me for the
period of Dec-2008 to March-2013 in the School/ Center/
Department of Computer Science & Engineering, Institute
of Engineering & Technology (IET), a constituent college of
Gautam Buddh Technical University, Lucknow.
Prof. (Dr.) S.P. Tripathi
(Supervisor)
CONTENTS
Acknowledgement i
Abstract iii
List of Figures v
List of Tables vi
List of Publications vii
INTRODUCTION 1
Preamble ………………………………………………………………….. 1
Need for Present Work …………………………………………………..... 1
Scope of the Present Work……………………………………..………….. 2
Objective of the Work …………………………………………………….. 3
Methodologies………………………………………………………………4
Organization of the Thesis…………………………………………………. 5
CHAPTER 1 LITERATURE SURVEY 8
CHAPTER 2 E-COMMERCE: INDIAN SCENARIO 15
2.0 Introduction……………………………………………………………. 15
2.1 E-commerce Unleashed………….……………………………………. 17
2.1.1 Types of E-commerce …………………………………….. 22
2.1.1.1 Business to Business (B2B)……………………….. 22
2.1.1.2 Business to Consumer (B2C)……………..……….. 22
2.1.1.3 Business to Government (B2G)…………..……….. 23
2.1.1.4 Consumer to Consumer (C2C)……………………. 24
2.1.1.5 Mobile Commerce (B2B)………………………… 24
2.2 E-commerce in India………….……………………………………….. 26
2.3 Top E-commerce Websites in India………….………………………… 32
2.4 Conclusion …………………………………………………………….. 38
CHAPTER 3 TRUST IN E-COMMERCE 39
3.0 Introduction ……. ………………………………………………………39
3.1 Defining Trust …………………….…………………………………… 40
3.2 Trust in E-commerce ………………………………………………….. 43
3.3 Current Approaches for Evaluating Trust…………………..………….. 45
3.3.1 Self-Regulation……………. ………………………..….…… 45
3.3.2 Presentation of Website …………………………..…….…… 47
3.3.3 Agent Based User Technology………………………….…… 48
3.3.4 Mathematical Approaches ………………………..…….…… 49
3.3.4 Public key Infrastructure and Security………….……….…… 49
3.3 Fuzzy Regression ………………………………………………………50
3.4 Trust Attributes ……………………………………………………….. 53
3.4.1 Web Object Content (WOC) ………………………………… 53
3.4.2 Web Object Ownership (WOO) ……………………………... 53
3.4.3 Web Object Certification Authority (WOCA)………………. 53
3.5 TrFRA Model ……………………………………………………….. 54
3.5.1 Brief Overview of Tanaka Model …………………………… 54
3.5.2 Case of Lucknow City with Trust Rating on eBay ………….. 55
3.6 Conclusion……………………………………………………………... 56
CHAPTER 4 TRUST V/S COMPLEXITY OF E-COMMERCE SITES 58
4.0 Introduction………………………………………………………. ……58
4.1 Trust Issues in Ecommerce………………………………………. …… 60
4.2 Existing Web Based Trust Models…………………………………….. 63
4.3 Our Proposed Model…………………………………………………… 65
4.4 Conclusion …………………………………………………………….. 68
CHAPTER 5 WEB PERSONILAZTION AND RECOMMENDATION 70
5.0 Introduction……………………………………………………………. 70
5.1 Web Personalization and Recommendation System…………………… 71
5.1.1 Types of Recommendation System………………………….. 75
5.1.1.1 Content-Based Filtering……………………………. 75
5.1.1.2 Collaborative Filtering……..………………………. 76
5.1.1.3 Knowledge Based Recommendation………………. 77
5.1.1.4 Utility Based Recommendation……………………. 78
5.1.1.5 Demographic Recommendation…………………….78
5.2 Our Proposed Model…………………………………………………… 79
5.3 Naïve Bayes……………………………………………………………. 79
5.3.1 Advantages & Disadvantages of Naïve Bayes Classifier……. 84
5.3.2 Fuzzy Information Classification and Retrieval Model……… 84
5.4 Conclusion………………………………………………....................... 89
CHAPTER 6 CONCLUSION 90
REFERENCES 94
ACKNOWLEDGEMENT
I am highly indebted to Dr. S.P. Tripathi for being my supervisor
and giving invaluable comments on my thesis. I have no words in praise
of Dr. R.P. Agarwal without whose able guidance I would have never
completed my endeavor. They have helped me with wise advice, useful
discussions, comments and facilities. They have never hesitated to
spend anytime or effort to guide my work. They helped me to overcome
the various difficulties and challenges that raised at various stages of
the thesis. It is just ineffable to express my deep gratitude towards him.
Special thanks to Dr. J.B. Singh who initially was my supervisor
and motivated me to carry my work forward.
Thanks to every single friend for his or her encouragement and
care to complete my thesis. I would like to thank everybody for their
blessings and support in this endeavor. Special thanks to Mr. Mahesh
Agarwal, (Managing Director), Goel Group of Institutions, Lucknow for
supporting me both financially and mentally to help me in timely
complete my work.
I cannot say enough “thank you” to my family for their
unconditional love and strength. My children Prateek and Juhi
patience and encouragement made this thesis possible. My wife
Meghna for supporting me at every moment in life.
i
I feel a deep sense of gratitude for my parents Mr. Hariom
Agarwal and Smt. Shashi Agarwal who formed part of my vision and
taught me the good things that really matter in life. The happy memory
of my family always provides a persistent inspiration for my journey in
this life.
I am grateful to my brother Amit & sister Pooja, in-laws, and all
relatives, for rendering me the sense and the value of strong bond with
them.
(Devendera Agarwal)
ii
ABSTRACT
In this era of digital world, Internet has emerged as a key business
medium of exchange and more and more electronic transaction are
being performed on it. The rise of the Internet has introduced many
kinds of virtual worlds where we interact with identities that conceal the
individuals behind. This limits our abilities to judge people in the same
way as we do in real life, therefore assessing another’s attitude and
character has become more difficult. In addition, members in virtual
communities can come from anywhere in the world, which considerably
lowers the likelihood meeting somebody we already know.
It has also been found that although there are many online services
providers' with websites in operation for some years, eCommerce has
still not achieved its potential. Some websites have been successful in
drawing high volumes of traffic; Web users were still reluctant to make
purchases from these websites. The trustworthiness of website is
increasingly affecting the rate of growth of eCommerce.
This problem being faced all over world we study its impact in full
deployment of Business-to-Consumer (B2C) e-commerce in India and
development of trust on the side of consumer. Consumer’s trust
concerns appear to be related to a number of factors, including
security, privacy, unfamiliarity, distance in time & space and unreliable
information.
iii
This thesis addresses to these issues with Indian Perspective and
highlight various Trust building factors both from the perspective of
buyers and sellers.
***
iv
LIST OF FIGURES
Figure Nos. Chapters Page Nos.
Chapter – 2 Figure. 2.1 Sub Categories of Websites 29
Figure. 2.2 Sample Acceptance Region 35
Figure. 2.3 Regression Line Showing Relationship between
Category & Prices
37
Chapter – 3
Figure. 3.l Membership Function of Fuzzy Coefficient ~A 52
Figure. 3.2 Fuzzy Output Function 54
Chapter – 4
Figure. 4.1 Trust/ Complexity Matrix 59
Figure. 4.2 Three Takagi-Sugeno Rules 67
Figure. 4.3 Input Output Data using Fuzzy Data Set 68
Chapter – 5
Figure. 5.l Indiatimes Shopping Portal 85
Figure. 5.2 Venn Diagram Illustrating Electronic Items with
Laptops as one of Categories & their Overlap
86
Figure. 5.3 Computing the Overlap 87
Figure. 5.4 Implementation Framework 88
v
LIST OF TABLES
Table Nos. Chapters Page Nos.
Chapter – 2
Table 2.1 Surfing Pattern in India 28
Table 2.2 Ranking & Percentage of Indians visiting these
Websites
33
Table 2.3 Contingency Table from the Data of Table 2.2 34
Table 2.4 Contingency Table from the Data of Table 2.3 35
Table 2.5 Comparative Prices of various Products Offered by
Ecommerce Websites
36
Chapter – 3
Table 3.1 Fuzzy Regression Value of Sample Data 56
Chapter – 5
Table 5.1 Report of Customer on Ecommerce Site 79
Table 5.2 Sample of 10 Users for fraudulent Intentions 82
vi
LIST OF PUBLICATIONS
1. Web Personalization of Indian e-Commerce Websites Using
Classification Methodologies, International Journal of
Computer Science Issues (IJCSI), Volume 7, Issue 6, 2010.
2. TrFRA: a Trust Based Fuzzy Regression Analysis,
International Review on Computers and Software (I.RE.CO.S),
Vol. 5 N. 6, 2010.
3. Trust Vs Complexity of E-Commerce Sites, International
Journal of Scientific & Engineering Research (IJSER), Volume 3,
Issue 4, 2012.
4. E-Commerce: True Indian Picture, Journal of Advances in
Information Technology, (JAIT), Vol.3, N. 4, 2012.
vii
INTRODUCTION
INTRODUCTION
Preamble
The present work proposes the confidence building measures that will
bring Indian masses to really trust and embrace Business to
Consumer (B2C) ecommerce. It proposes a framework for measuring
Trust on a website and also derives relationship between Trust and
Complexity. It also highlights the importance of Web Personalization
and Recommendation System with help of the Classification
Methodologies using Bayesian Rule and predicting fraudulent
transaction.
Need for Present Work
India is a vast country with huge population that makes it an
interesting prospect for market researchers to launch new products &
technology. India has always embraced them with open mind.
However after the much hyped dot com disaster in year 2000,
ecommerce has taken a huge setback. Ecommerce is the perfect
technology for Indians but there are various reasons inhibiting them
from accepting it the way we have embraced other technologies. Tall
claims have been made, hypothetical sales figures have been
displayed but all are far from truth. Ecommerce for masses is
Business to Consumer ecommerce where a vendor or companies want
to sell a product to End-Consumer. In other countries high
PhD. Thesis by Devendera Agarwal 1
percentages of users are shopping online, in India this percentage is
timid as compared to internet users with other countries.
Ecommerce which offers so many advantages and saves lot of time in
populous country like India, it becomes essential to ascertain reasons
inhibiting consumers from crossing over the last hurdle i.e carrying
out the monetary transaction.
Scope of the Present Work
The scope of the present work is divided into various phases:
1. Highlighting reasons keeping away consumers from using e-
commerce.
2. Highlighting the significance of Trust in e-commerce, identifying
Trust Attributes and modeling a tool TrFRA for measuring Trust
of a website.
3. Highlighting the relationship between Trust and Complexity, so
as while increasing Trust we may not ruin the overall experience
of shopping and drive away consumer from the website.
4. Highlighting the importance of Web Personalization and
Recommendation System in increasing the sales of a website at
the same time predicting the fraudulent transaction to
safeguard the interests both consumer and vendor.
PhD. Thesis by Devendera Agarwal 2
Objective of the Work
The objective of the present work is:
a) To build up considerable knowledge on factors enhancing
consumer trust in e-commerce websites.
Our first objective is to identify various factors likely to
influence people trust in an e-commerce website. Having identified
these factors, we will analyze and classify them into suitable groups to
form a model of trust in e-commerce. During our research, we will
concentrate on the issue regarding of what factors makes consumer
trust one website more than another, while believing that our target
audiences have the essential basic trust to transact online.
b) To build up and endorse web personalization and recommend
a system in accordance to Indian e-commerce websites.
The second objective of this research is to propose web
personalization and recommend a system as per shopping behavior of
Indian consumers. We have adopted several techniques, in order to
generate a hybrid self-learning recommendation system. With this
system, the customers can very likely find the most suitable products
in the shortest time without spending much effort.
c) To build up and substantiate procedural knowledge to devise
and assess trust enhancing factors in e-commerce websites.
The last goal is to use this model of trust from previous phase
and propose a tool to help website developers evaluate website
trustworthiness, as well as developing a site with focus on trust
building factors which finally leads to pleasant shopping and improves
PhD. Thesis by Devendera Agarwal 3
overall virtual shopping experience safeguarding both vendor and
consumer by predicting fraudulent transaction.
Methodologies
Web personalization, consist of activities such as providing
customized information, changing the webpage layouts and adapting
the contents tailored to the user’s need. It has become an essential
part of a website to enhance its compatibility and attractiveness.
The recommendation system, a technology that enables the e-
commerce websites to predict a user’s interest for a product and
provide personal shopping assistance according to individual
background information, preferences and needs, have become a
reliable and influential tool for the customers [Senecal & Nantel,
2004]. Evidently, more and more e-commerce sites resort to
recommendation systems to interact with the customers and hopefully
increase sales.
Human Computer Interaction (HCI) with its pedigree in psychology
and ergonomics has been defined as [ACM SIGCHI, 1992]
“the discipline concerned with the design, evaluation, and
implementation of interactive computing systems for human use
and with the study of major phenomena surrounding them.”.
The major advantage of this methodology for this research is its ability
to gather user opinion about particular factors systematically and to
ascertain whether they have any directly relation to the site’s design
or not, its influence in acceptance and usage, and, therefore without
reservation, user trust in that website. Meanwhile, the major
PhD. Thesis by Devendera Agarwal 4
constraint of this approach is that we deal only with the subjective
reports about apparent trustworthiness and as a result we will not
exactly deal with main risk and trustworthiness.
Even though the methodology adopted for the research is highly
academic in nature but once the research is complete it will benefit
dot com industry in India and will find great acceptance.
Organization of the Thesis
Chapter 1 presents literature review of the E-commerce scenario in
India, factors responsible for growth of Computers, Internet and E-
commerce in India. Having identified the factors inhibiting the user
from carrying out transaction over internet i.e. Security, Trust,
Complexity and Uncertainty. To overcome these anomalies it highlights
the work of P. Luansal and Diego Gambetta for Trust, H. Ishibuchi &
H. Tanaka for Uncertainty, work of M. Setnes, T. Takagi and M.
Sugeno for Complexity and lastly work of J.B. Schafer, Chung Chen
and L Chen for Security.
Chapter 2 starts with the brief description of India’s rich heritage and
economy in comparison with other developing and developed
countries. It highlights various reasons responsible for sudden surge
in Computers and Internet in India over last decade. It defines and
gives classification of e-commerce before focusing on surfing pattern
of Indians and the scenario of e-commerce in India. It explains in
detail What and Why Indians are lured on the net. It highlights the
PhD. Thesis by Devendera Agarwal 5
relationship between Indian Rank and World Rank of Indian e-
commerce websites and concludes with future of e-commerce in India.
Chapter 3 contains the role of Trust in e-commerce. It defines Trust
in virtual world and how it differs from Trust in real world and also
discusses the importance of Trust in B2C e-commerce. Later in the
chapter it focuses on relationship of Trust in Indian e-commerce sites
with special reference to three factors: Web Object Content, Web
Object Ownership and Web Object Classification Authority. It
concentrate on Fuzzy Relationship between Trust and these three
factors using Fuzzy Regression Analysis and concludes with TrFRA
model for measuring Trust on Indian e-Commerce Websites.
Chapter 4 is an extension of work in previous chapter. It focuses on
that while increasing Trust one usually ends up increasing the
Complexity of an E-commerce Transaction thus driving away the
consumer from the site. It defines a Fuzzy relationship between Trust
and Complexity using Three Takagi-Sugeno Rules and Tanaka Model.
Chapter 5 deals with the Security and the Web Personalization issues
of a website. It discusses the importance of Web Personalization and
Recommender System on an E-Commerce website and also how it
helps in understanding Consumer Behavior thus increasing the sales
of a website. It then highlights the use of Classification Methodologies
along with Bayesian Rule for Indian e-commerce websites. It deals
with generating cluster of user having Fraudulent Intentions and also
focuses on Bayesian Ontology Requirements for efficient Possibilistic
Outcomes.
PhD. Thesis by Devendera Agarwal 6
Chapter 6 draws the previous chapters to a conclusion and indicates
the scope of the Trust, Web Personalization and Recommendation
System, how and where it should be applied and the potential future
developments of the following in Indian ecommerce websites.
It sums up an approach to improve the overall shopping experience of
a consumer on the net by increasing his Trust and keeping
Complexity within limits so as not to drive him away to another
website.
***
PhD. Thesis by Devendera Agarwal 7
CHAPTER – 1
LITERATURE SURVEY
Chapter – 1
LITERATURE SURVEY
The thesis focuses on Trust, Web Personalization and Recommender
System to build upon the confidence of perspective buyer browsing
websites for products and yet restraining them to cross the final
hurdle of carrying out the transaction for some unspecified reasons.
The thesis works in two phases. First phase is the survey part, which
emphasize on the present scenario of Computers, Internet and E-
Commerce in India. Computers and Internet being inherent to E-
commerce are essential for growth of B2C ecommerce. Second phase
an extension of first phase proposes various methods to overcome
problems identified to increase Trust and Sales of B2C ecommerce
sites.
First phase starts with study of various surveys mainly conducted by
IAMAI (Internet and Mobile Association of India) a non profitable
company to enhance and expand the online and mobile services in
India and IMRB (Internet Market Research Bureau) International a
leading market research and survey Company. They carry survey in
conjunction pertaining to Mobile, Internet and E-Commerce. Their
surveys “Consumer E-commerce Market in India 2006/07”, “Online
Activities & Ecommerce from Cybercafés 2005”, “ICUBE 2007” and
“ICUBE 2008” discusses various Triggers and Barriers for Ecommerce
market in India, these survey also suggest that maximum 10% of
PhD. Thesis by Devendera Agarwal 8
users actually carried out transaction and that too for non-tangible
products while in case of tangible products it falls to merely 6%, while
biggest contributor being online travel industry.
IOAI (Internet and Online Association of India) survey “Ecommerce
Security 2005” is an executive summary providing an insight about
Internet Access & Usage details, Security of Online Transactions,
Areas of Concern while shopping online and also factors that would
increase the faith in Online Transaction.
KPMG Forensic, India surveys “India Fraud Survey Report 2006” and
“India Fraud Survey Report 2008” gives an insight that e-fraud even
though are rare but dents the Trust and Confidence of consumer and
drives him away from internet.
PEW Internet & American Life Project survey “Online Shopping” in 2008
carried out this survey in America and found that 65% internet users
have purchased online and also find various problems faced by others
that prevented them from carrying online transaction. To our surprise
biggest reason being lack of Trust.
INTEL-IMRB study “Indian Urban Consumer Segment Nationwide Study
2009-10”, discuss about the rising PC’s penetration in urban homes
in Chapter-2 we have discussed reason for this sudden surge in
demand of PC’s.
IAMI-IMRB studies “45 Million Internet Users in India” in 2010, “84%
rural India in unaware of Internet” in 2010 and “India will have 5.4
million rural internet users” in 2008 provide an insight how internet
PhD. Thesis by Devendera Agarwal 9
usage in India is growing, what is potential for growth and from where
Indians are accessing the internet.
Study of all these surveys helps us to determine factors for PC
penetration, Rise in Internet usage and e-commerce in India, Surfing
Pattern, Internet access place, Buying Pattern of Products over the
Internet and determine factors inhibiting Indians from carrying out
online transactions.
Second Phase starts with Trust and Security as key ingredients, the
conceptual understanding of the meaning of trust explores different
types and aspects of trust, including both internal and external
organizational trust [75], the theory of trust [39, 95, 105] and beliefs-
based trust [1, 7]. For example, Belanger et al have used Diffusion of
Innovation Theory and literature on trustworthiness in eCommerce as
theoretical bases. From a social perspective, Friedman et al believe
“people trust people, not technology”. They also suggest that online
trust can be cultivated through ten trust related attributes of online
interactions, such as security and reliability of technology; knowing
what people end to do online; absence of deceptive content and
pictures; Conflict about what counts as harm; informed consent;
anonymity; accountability; importance of sign in the online
environment; insurance & functioning history and repute. They point
out that “we are vulnerable to trust violations in two ways: loss of
money and loss of privacy”.
PhD. Thesis by Devendera Agarwal 10
A number of conceptual models for online trust have been proposed
[3, 33, 76, 86]. For example, Tan et al proposed a business-to-
business trust model using agent-based technology in online
transactions. Cheskin eCommerce Trust Study explains that
trustworthiness is a function of time and certain formal
characteristics of a websites. Web users depend upon certain methods
being followed. Over the period of time, dependence on these methods
gives way to dependence on experience. Such dependence is
prerequisite for true trust to develop. Cheskin’s study identified the
following fundamental ‘forms’ that communicate trustworthiness to
web users, including Brand, Presentation, Fulfillment, Navigation,
latest Technology and the symbol of security guaranteeing firms.
Other studies [2, 16, 19, 23, 29, 86, 87, 93] have also identified
additional trustworthiness factors including seals of approval, ease
navigation, presentation, and technology as contributing to Web trust.
Seals of Approval: Information about dedicated companies that
concentrate in promising the safety of websites (symbols like
Verisign). Seals of approval try to comfort the visitors that
control has been instituted.
Brand: Company promise in delivering precise attributes and
their integrity, based upon reputation and users possible past
experiences. Companies with an established and respected
offline presence can control their brand equity in an online
environment. Such ‘click-and-mortar’ vendors are often
PhD. Thesis by Devendera Agarwal 11
supposed to be more legitimate and trustworthy than their pure
online competitors [86].
Navigation: The ease in finding what visitor seeks through
quickly, manuals and instructions to help the user to perform
search and transaction tasks on the website [96].
Assurances: Online consumers need to feel that the information
given by them will be kept secure and will only be used for pre-
specified purposes. Online vendors, often with the help of third
parties, must provide security and privacy assurances in order
to facilitate a trusting relationship with their consumers [65, 86,
96].
Fulfillment: A clear indication of functionality like how orders
will be processed and how problems will be addressed. As in the
offline environment, online vendors must ensure their promises
are fulfilled to establish and maintain trust and long-term
relationships with consumers.
Website Presentation: Design aspect that leads to quality and
professionalism. In an online environment, the vendor's website
is the primary interaction mode for its customers. The design
and usability of the website can predict user perceptions of
trust. For example, sites that are easy to navigate, easy to
search, professionally designed, well networked, informative,
timely and accurate can greatly influence an online consumer's
perception of trust [65, 86].
PhD. Thesis by Devendera Agarwal 12
Technology: State of the art technology connotes
professionalism, even if it is difficult to use, i.e. website
technical functions. For example, how quickly pages, text and
images appear and how well the website seems to work overall
[104].
Head et al proposed a theoretical framework to build trust in
eCommerce, “which identifies the primary trust parties and their
interactions throughout the trust lifecycle, which is a dynamic process
that deepens or retreats based on experience”. These parties are
“Consumers who seek trust before transacting with a vendor; vendors
who seek to build trust among consumers; and Referees who provide
independent recommendations on the trustworthiness of vendors”.
We are inspired by work of P. Lamnsal [89] who explain Trust and
Security, Diego Gambetta [26] who describes whom to Trust and
whom Not to Trust and also Chervany et al.[85] and as inspiration
Josang et al. [63, 64] describes the meaning of Trust and also the
semantic constraints for trust transitivity, L.J. Camp[14] work on the
threats to ecommerce and Koh Ai Tee [66] legitimacy condition for
development of Trust and Yinan Yang et al. [127] work on identifying
set of trust attributes helped me in the development of TrFRA model
for measuring Trust on a website.
While extending the model to include complexity with Trust we are
inspired by the work of H. Ishibuchi & H. Tanaka [43, 44] who
highlights the construction of Fuzzy Classification of various entities
using genetic Algorithms. Later on they extended their work (1995)
PhD. Thesis by Devendera Agarwal 13
using “If-Then-Else rules”, M. Setnes [81] developed a Rule-Based
system for developing the Precision & Transparency. D. Nauck [28]
worked on the interpretability aspect of Medical Data; motivates us for
extending it for e-commerce websites. Y. Jin [124] developed a
framework for modeling high dimensional system in finding out their
Complexity and Interpretability aspect. L. Castillo [71] developed the
best rule in a genetic fuzzy learning algorithm. M. Setnes [80] also
developed a mechanism of GA-based Modeling & Classification which
measures the Complexity and Performance of the system. T. Takagi &
M. Sugeno [109] helped us to derive a model which helps in finding
out trading off between Trust and Complexity.
For recommendation system we are motivated by work of J.B. Schafer
[108, 110] on application of recommender system and later extension
of work in [109] using data mining and knowledge discovery. Work by
Ming-Chung Chen [19] and L. Chen [17] for mining user behavior for
ecommerce using virtual reality technique and Lipo Wang [121] work
on fuzzy system for knowledge discovery and of the models by G.
Haubl et al.[46], D. Menom Smith et al.[112] & S. Senecel et al.[106]
work on consumer decision making and influence of online
recommendation on decision making also helped us in realizing
Indian consumers perspective in online purchase.
***
PhD. Thesis by Devendera Agarwal 14
CHAPTER – 2
E-COMMERCE: INDIAN SCENARIO
Chapter – 2
E-COMMERCE: INDIAN SCENARIO
2.0 Introduction
India is a country with rich historical heritage, the second most
populous country and the most populous democracy in the world. It
has achieved multifaceted socio-economic progress during the last 63
years of independence and has once again emerged on world scenario
as one of largest economies.
India [123] primarily being country whose economy encompasses the
traditional village farming, finally accepted computer as an ally not
foe. Computer growth in India moved in leap and bounds after much
hyped Y2K problem. Since then India have make its presence felt
worldwide in software industry with companies like Infosys, TCS, and
Wipro etc. grew exponentially. According to the International Monetary
Fund, India's nominal GDP stood at US$1.3 trillion, which makes it
the eleventh-largest economy in the world, corresponding to a per
capita income of US$1,000. If purchasing power parity (PPP) is taken
into account, India's economy is the fourth largest in the world at
US$3.6 trillion. The country ranks 142nd in nominal GDP per capita
and 127th in GDP per capita at PPP. With an average annual GDP
growth rate of 5.8% for the past two decades, India is one of the
fastest growing economies in the world.
PhD. Thesis by Devendera Agarwal 15
According to 2011 PwC report [97], India's GDP will go beyond that of
Japan in 2011 and by 2045, India's GDP will be better than that of
the United States. Moreover, over the next four decades, in India's
average annual economic growth rate at about 8% is expected to
stand and therefore, India has the potential to be one of world's
fastest growing major economy over the period to 2050. India has
large numbers of well-educated people skilled in English language;
India is a major exporter of software services and software workers.
The Indian Information Technology industry accounts for 5.18% of the
country's GDP and export earnings as of 2009, also providing
employment to a majority of its service sector workforce. More than
2.3 million persons are employed in the sector either directly or
indirectly, thus making it one of the biggest job creators in India and
an important pillar of the national economy.
Privatization of technical education also took place during this period
and today India which churns out almost a million engineers every
year giving boost to IT growth. Public schools are not too far behind in
imparting computer education as it was made compulsory from IIIrd
standard itself. According to a report, with title “India Urban
consumer segment nationwide study 2009-10”, surveyed 19,178
respondents across 82 cities by Intel and IMRB, it was reported that
computer penetration in urban India doubled in last three years from
19 per cent to 38 per cent and now nearly 28 million households have
the PC in their houses.
PhD. Thesis by Devendera Agarwal 16
The PC purchases have been driven by better education opportunity,
internet connectivity and ease of working from home. Multipurpose
usage of PC for gaming, watching videos and listening to music has
also kicked off the sales of PC.
Technical education in India is governed by All India Council of
Technical Education (AICTE) [5], which makes it compulsory to
maintain ratio of 1 computer for every 4 seats in an engineering
institute. It is also compulsory to have dedicated internet connectivity
with bandwidth of 1 Mbps or more. Under normal circumstances any
institute running for over 4 years must have at least 300 PCs.
Moreover more and more institutes are offering students laptops at
subsidized rates. School children have also started to demand
computers for practice at home.
This gives an insight to why computer sales have surged in last few
years. Computer and Internet are two essential components for e-
commerce. Their increase has certainly had a positive impact on e-
commerce growth in India.
2.1 E-Commerce Unleashed
The electronic commerce concept was developed in the 70's, even
though electronic commerce under the infant form of EDI or electronic
data interchange has been existing since the late 60's with the
invention of the first data networks [103].
Electronic commerce, also called as e-commerce or eCommerce,
comprises of the buying and selling of products and services over
PhD. Thesis by Devendera Agarwal 17
electronic systems like Internet and other mediums. Total trade
carried out electronically has grown enormously with extensive use of
Internet. The use of commerce is accomplished in this way is
encouraging and improvement is visible in transferring funds,
managing supply chain, internet marketing, transaction processing,
electronic data interchange, inventory management systems, and
automatic data collection systems. Now a day’s electronic commerce
typically uses World Wide Web at a point in the transaction's life cycle,
even though it can cover a wider range of technologies including e-
mail as well.
Major portion of electronic commerce is conducted electronically for
non tangible items such as access to paid content on a website, but
remaining electronic commerce involves the shipping of physical items
in some way. Online retailers sometimes also known as e-tailers and
online retail is known as e-tail. Now a day’s almost all big retail
houses have their presence electronically on the World Wide Web.
Electronic commerce is by and large considered to be the sales portion
of e-business. It also consists of the swapping of data to facilitate the
financing and payment portion of the business transactions.
Originally, electronic commerce which is known for making electronic
commercial transactions easy, using technologies such as Electronic
Data Interchange and Electronic Funds Transfer. Both being
introduced sometime during the late 1970s, permitting businesses to
send various documents like purchase orders or invoices
electronically. Increase in growth and world wide acceptance of credit
PhD. Thesis by Devendera Agarwal 18
cards, automated teller machines and telephone banking in the 1980s
may also be termed as form of electronic commerce.
In 1990, Tim Berners-Lee invented the World Wide Web browser and
transformed an academic tele-communication network into a world
wide everyman communication system called internet / WWW.
Commercial ventures on the Internet were strictly illegal until 1991.
Although the Internet gained popularity worldwide around 1994 when
online shopping first took place on internet, it took another five years
to implement security protocols and DSL permitting incessant
connection to the Internet. By millennium end, various business
companies in America and Europe started offering their services over
internet. Since then people have started associating a word "e-
commerce" with the ability of purchasing goods over the Internet using
secure protocols and electronic payments.
With the advent of the World Wide Web (WWW), electronic commerce
and especially company-to-consumer electronic commerce, is based
on public networks such as Internet. Their main characteristic being
that they are less expensive and widely accessible not only by
corporations but also by the single individuals. There are many
definitions of electronic commerce and much confusion there is about
this term.
PhD. Thesis by Devendera Agarwal 19
For example Wigand [122] states that:
“Electronic commerce denotes the seamless application of
information and communication technology from its point of origin
to its endpoint along the entire value chain of business processes
conducted electronically and designed to enable the
accomplishment of a business goal. These processes may be
partial or complete and may encompass business-to-business as
well as business to consumer and consumer-to-business
transactions”
Zwass [129] defines electronic commerce as:
“The sharing of business information, maintaining business
relationships, and conducting business transactions by means of
telecommunications networks…Therefore as understood here, E-
commerce includes the sell-buy relationships and transactions
between companies, as well as the corporate processes that
support the commerce within individual firms”
A broader definition by Kalakota and Whinston [102] is:
“E-commerce is associated with the buying and selling of
information, products and services via computer networks today
and in the future via any one of the myriad of networks that
make up the Information Superhighway (I-way)”
PhD. Thesis by Devendera Agarwal 20
Internet also enables the marketers to easily reach the customers and
promote their brands or products by offering vast product information
and options. Electronic Commerce is about buying and selling of
goods and services electronically by consumers or by companies via
computerized transactions. E-Commerce has speeded up ordering,
production, delivering, payment for goods and services by replacing
manual and paper based business processes with electronic
alternatives and by using information flow effectively in new and
dynamic ways. At the same time, e-Commerce has reduced marketing,
operational, production, and inventory costs in such a way that
customer will also benefit indirectly.
Therefore, Internet is the technology [37, 48, 94] for e-Commerce as it
offers easier ways to access companies and individuals at a very low
cost in order to carry out day-to-day business transactions. Around
the clock presence of companies on the Web gives competitive
advantage to companies’ businesses.
However, since the Internet is publicly accessible, data can be more
easily intercepted, which seriously undermines the security of online
transactions, as well as the privacy and confidentiality of the
commercial exchange.
Moreover, the legitimacy and the trustworthiness of online vendors
cannot be guaranteed as adequately as on a private network, because
there is no control as to who will enter the system and how parties will
authenticate themselves. Since users will often have the choice
between a large numbers of different business partners and since the
PhD. Thesis by Devendera Agarwal 21
cost of switching from one vendor to another is negligible, it is
imperative that online vendors stand out by addressing not only users’
functional business needs, but also their concerns in terms of
security, confidentiality and trustworthiness.
For private users to adopt e-commerce, it is very important that the
benefits of using the new commercial medium (e.g. convenience of
home, decreased transaction costs, time etc.) significantly over
shadow potential risks. Private user's freedom to choose proper
vendors is likely to be correlated with greater concerns with regard to
financial risk, privacy and trust. This can also be accounted for by the
fact that private users are more directly drawn in the commercial
exchange, since they are using their own equipment, giving sensitive
information about themselves as individuals, and spending their own
money.
2.1.1 Types of E-Commerce
The different types of e-commerce are described below in brief:
2.1.1.1 Business-To-Business (B2B)
Business-to-business or B2B e-commerce is simply defined as
e-commerce between companies. This type of e-commerce deals
with relationships between and among businesses. About 80%
of e-commerce fall under this category, and many experts are of
view that B2B ecommerce will grow faster than B2C segment.
2.1.1.2 Business-To-Consumer (B2C)
Business-to-consumer e-commerce, is commerce between
companies and consumers, it involves customers collecting
PhD. Thesis by Devendera Agarwal 22
information; buying physical goods (i.e., tangibles such as
phones or consumer products) or information goods (like
digitized content, such as software, songs or e-books); where
information goods are mainly received over internet.
It is second largest and the primitive form of e-commerce. Its
genesis can be traced to online retailing. Thus, it being more
common B2C business models is the online retailing, various
companies such as Amazon.com, Drugstore.com, Yahoo.com,
Rediff.com and Indiatimes.com fall under this category.
B2C e-commerce cut transactions costs (especially search costs)
by increasing the consumer access to information and
permitting consumers to find the best price for an item or
service. B2C e-commerce also reduces market entry barricades
since the cost of setting up and maintaining a website is
considerably much cheaper than installing and maintaining a
“brick-and-mortar” structure. In case of information supplies,
B2C e-commerce is even more lucrative because it saves firms
extra cost of a physical distribution network. The country with
an ever growing and haelthy Internet population, like India
delivering information goods becomes increasingly feasible.
2.1.1.3 Business-To-Government (B2G)
Business-to-government or B2G e-commerce is generally
defined as commerce between companies and public sector. It
means using the Internet for licensing procedures, public
procurement and other government-related operations.
PhD. Thesis by Devendera Agarwal 23
A web-based policy for purchasing increases the transparency
in the procurement process and also reduces the risk of
irregularities. However till date the contribution of the B2G
ecommerce in total e-commerce is very less, as government e-
procurement systems mainly remains undeveloped.
2.1.1.4 Consumer-To-Consumer (C2C)
Consumer-to-consumer or C2C e-commerce is basically
commerce between private individuals. This type of e-commerce
can be attributed to the growth of electronic market places and
online auctions, predominantly in industries where firms or
businesses can bid for the required product from among
multiple suppliers.
2.1.1.5 Mobile Commerce (m-commerce)
M-commerce is simply buying and selling of merchandise and
services using wireless technology i.e., hand held devices like
cell phones and personal digital assistants (PDAs).
As carrying data over wireless devices is becoming much faster,
more secure, and scalable, many have started to believe that m-
commerce will soon overtake wire line e-commerce as method of
choice for digital e-commerce transactions. This surely is
certainly true for the Asia-Pacific region where more persons
prefer mobile phone rather then computer for Internet.
This brief write-up with e-commerce discusses the evolution of e-
commerce and how it has become an almost necessity in our day to
PhD. Thesis by Devendera Agarwal 24
day life. Frequent developments in technology particularly 3G and 4G
in mobile will only add to the speed of growth of e-commerce.
Our major emphasis will be on B2C, as this type e-commerce is e-
retailing or more common e-tailing. It involves the process of billing
the end consumer. In our view this is where the true test of e-
commerce takes place.
E-commerce requires monetary transaction, one single step where
user hesitates to complete transaction. He already has heard so many
electronic frauds [56, 67], computers being hacked, passwords stolen
etc., on the contrary truth is only (0.03%) of B2C transaction are
fraud. 87% of fraud comes from online auctions (C2C e-commerce). If
this myth can be broken it will prove to be a big leap in e-commerce.
In B2B and B2G e-commerce one party is a business house while
other being a business house or a government organization. Both of
them are well aware of threats of e-frauds which include manipulation
of data records, hacking into organization systems, manipulation
computer programs, unauthorized transfer of funds, failure of an e-
transactions etc. Both business houses and government organizations
have cyber security cells to maintain their computer system, networks
with firewalls in place, proxy servers, antivirus software, and white-list
authorized wireless connections etc. They also have facility of legal
advice to handle cyber crimes. On other hand a consumer in very
small fish in sea who wants cheapest products usually falls in trap
knowingly or unknowingly.
PhD. Thesis by Devendera Agarwal 25
An e-commerce transaction requires a PC with internet connectivity
and can be carried out from Home, Cybercafé or Office. At home we
can assume PC to be safe, at office we have proper security of routers,
ISA Server, Firewall and Antivirus so perhaps most secure place, while
a cybercafé is perhaps the most susceptible place for a fraud.
It is essential to safeguard the interest of the consumer; it is he who
will decide the fate of e-commerce in future. His trust has to be build;
e-commerce will automatically grow with his trust and confidence.
You can cheat him once, only to drive him away and not to trust e-
commerce.
2.2 E-Commerce in India
Tall claims have been made about internet usage and e-commerce [44,
93, 96, 120] in India. Let’s not go by the amount, as in B2B the
numbers of transactions are negligible but amount involved is huge.
B2B has always been here in form of EDI, so why there is so much
fuss. It is B2C and C2C e-commerce which constitute majority of
transactions of comparatively small amount.
A study, titled ‘India Urban Consumer Segment Nationwide Study
2009-2010 surveyed 19,178 respondents across 82 cities by Intel and
IMRB, it was reported that Computer use in urban India has just
doubled over last three years from 19 per cent to 38 per cent and now
nearly 28 million households have the PC in their houses. The study
also noted that youth in the age group of 18-25 have played an
important role as facilitators during actual purchase of computer.
PhD. Thesis by Devendera Agarwal 26
The PC purchases [101] have been driven by better education
opportunity, internet connectivity and ease of working from home.
Multipurpose usage of PC for Gaming, watching videos and listening
to music has also kicked off the sales of PC.
Study also found that, more first time buyers are purchasing laptops
as their first computer. In 2006, only a mere 17 per cent of first-time
buyers wanted to go for notebooks, while in 2009, the percentage of
non-owners who wanted a notebook instead of a desktop PC doubled
to 31 per cent.
One thing is sure that internet usage in India is increasing in leap and
bounds. Many surveys [49-55] show same is the case with e-
commerce. Lets not go by the amount, as in B2B the numbers of
transactions are negligible but amount involved is huge. It has always
been there in form of EDI, so why there is so much fuss. It is B2C or
C2C e-commerce where we have many transactions of small amount.
In this section first we find out the surfing pattern of Indians, to get
the answer to two primary questions.
a) Are internet users really interested in e-commerce?
If answer to above question is yes than to find
b) What we they buying on the internet and why?
We found out the top 100 websites (hit ratio) [4] in India and
classified them; the result is shown in Table-2.1.
PhD. Thesis by Devendera Agarwal 27
Categories Total
Portals 36
Advertisement 16
Entertainment 14
Social Networking 13
Search Engines 7
Internet Service Providers 6
Banks 4
E-Commerce 3
Encyclopedia 1
100
Table-2.1: Surfing Pattern in India
Now let’s explain each category and try to assimilate the necessary
information on surfing behavior, and If possible, find out shopping
pattern.
(A) Portals
A web portal, presents information from diverse sources in a
unified way. Apart from the standard search engine feature, web
portals offer other services such as e-mail, news, stock prices,
information, databases and entertainment.
When further sub-categorized the picture came out as shown in
figure-2.1. It is being observed that few portals are involved in e-
PhD. Thesis by Devendera Agarwal 28
commerce activities but keeping in mind Indian scenario, name
of six websites is worth mentioning i.e. (Rediff India, Indiatimes,
Ebay India & Sify) under General, Shaadi under Matrimonial
and Makemytrip under Travel Sub-categories. Remaining other
offer various other services and hence are not worth mentioning.
Fig 2.1: Sub Categories of Websites
(B) Advertisement
Under these categories are those websites whose names one
might have never heard of. They consist of those websites which
generally pop on your screen when you visit other websites.
PhD. Thesis by Devendera Agarwal 29
They are basically advertisements on other websites and simply
increasing their hit ratios Komli, Sulekha, Quikr India etc. are
few of them.
(C) Entertainment
This category is yet another extension of portals, where whole
emphasis is on entertainment only. It includes websites offering
(Games, Music, Videos, Cricket scores etc.)
(D) Social Networking
It consist of latest in fashion sites meant for communication
with friends, social causes etc. consist of common websites like
Facebook, Twitter, Orkut, Bharatstudent etc.
(E) Search Engines
Perhaps one of most powerful tool on internet for actual
working, no surprise, first two most popular websites being
search engines Google and Google India, other are Bing, Ask,
etc.
(F) Internet Service Providers (ISP)
Again like advertisement category it comprises of websites
which user actually doesn’t visit himself but their hitcounter
automatically hits when we open some other websites like
websites of Hit counter (StarCounter, Conduit, etc.) and Domain
Names at any spelling/typing mistake in name (GoDaddy,
DomainTools, etc.).
PhD. Thesis by Devendera Agarwal 30
(G) Banks
Banks are financial institutions an essential requirement for
carrying out monetary transactions in e-commerce. There are
actually three banks with one bank having two domain names
(HDFC, ICICI and SBI). HDFC and ICICI are largest private
banks in India; they are also pioneers of Internet Banking in
India. Any net savvy user will certainly have an account in these
banks. Third State Bank of India (SBI) is largest and the oldest
bank in India also offers net banking facility. Most business
organizations have account in it and again finding its name in
this category is no surprise.
Presence of these names suggests that e-commerce is certainly
present and making impact on India growth story.
(H) E-Commerce
Under these category we have placed those files that are strictly
e-commerce (B2C) websites (irctc.co.in & amazon.com) and one
being third party money transfer (paypal.com). Amazon is one of
world most popular e-commerce websites, it is more likely that
search engines direct user to Amazon rather user visiting this
site for buying a product since payment is in Dollars & not in
Rupees.
PayPal is again facing problems with RBI guidelines so its
presence in top 100 may be due to B2B transactions or some
other reason and not from B2C transactions. Website of Indian
Railway Catering & Tourism Corporation is a classical example
PhD. Thesis by Devendera Agarwal 31
of e-commerce growth in India and we discuss it in next section
in detail.
(I) Encyclopedia
The single website of Wikipedia must attribute its presence to
search engines, as it offers definition and history of each term
being searched and it one of automatic choice for visiting the
site.
In top 100 websites as per India’s choice and conditions seven
websites clearly can be termed as involved in B2C as they are selling
products or services to Indian public in their own currency, and four
websites providing payment gateway to carry out these transactions.
2.3 Top E-Commerce Websites in India
Our next step was to individually analyze top e-commerce websites
and find out what they have to offer and how much impact they make
on India e-commerce growth. In the Table-2.3 we have displayed these
websites with their Indian Rank, World Rank and percentage of Indian
audience visiting these websites.
PhD. Thesis by Devendera Agarwal 32
S.No Website Indian Rank
World
Rank
%Age Indian
Audience
1 Rediff India 9 145 89
2 Indiatimes 12 173 78
3 IRCTC 36 574 98
4 eBAY India 52 854 95
5 Sify 64 833 83
6 Shaadi 90 978 76
7 Makemytrip 98 1335 96
(Source: as per data available from alexa.com) Table-2.2: Ranking & %Age of Indians visiting these Websites
Here Rediff India, Indiatimes and Sify can be grouped together in B2C
category selling products of varied types viz. mobiles, laptops, camera,
watches etc., while eBay also deals in same but deals in C2C e-
commerce.
We will be checking whether there exists a perfect combination
between Indian Rank, World Rank and %age of Indian audience. To
check this we use chi-square test [22, 40] of independence,
considering the hypothesis as:
H0: PR1 = PR2
H1: PR1 PR2
Where
PR1 = Proportion of Indian Rank
PR2 = Proportion of World Rank
PhD. Thesis by Devendera Agarwal 33
Rediff
India
Times IRCTC
eBAY
India SIFY Shaadi
Makemy
Trip Total
Indian Rank 9 12 36 52 64 90 98 361
World Rank 145 173 574 854 833 978 1335 4892
Table-2.3: Contingency Table from the Data of Table 2.2
Combined proportion of Indian Rank = 361/4892
= 0.07
f0 fe D=f0-fe D*D D*D/fe
9 51.57 -42.57 1812.20 35.1407
12 51.57 -39.57 1565.78 30.3623
36 51.57 -15.57 242.42 4.7009
52 51.57 0.43 0.18 0.0036
64 51.57 12.43 154.50 2.9960
90 51.57 38.43 1476.86 28.6381
98 51.57 46.43 2155.74 41.8023
145 698.85 -553.85 306749.82 438.9351
173 698.85 -525.85 276518.22 395.6761
574 698.85 -124.85 15587.52 22.3045
854 698.85 155.15 24071.52 34.4445
833 698.85 134.15 17996.22 25.7512
978 698.85 279.15 77924.72 111.5042
1335 698.85 636.15 404686.82 579.0754
X2= 1751.3349
Table-2.4: Contingency Table from the Data of Table 2.3
PhD. Thesis by Devendera Agarwal 34
We compare the observed value of x2 with critical values of x2 and
apply the rules of hypothesis:
x2observed < x2critical => Accept the Null Hypothesis
and if
x2observed > x2critical => Reject the Null Hypothesis
Now calculation the degree of freedom [34, 117] we get:
F = (r – 1) (c – 1)
F = (2 – 1) (7 – 1)
F = (1) (6)
F = 6
at = 10% = 0.10
x2critical / = 0.10
Acceptance Region
X2 Observed
12.017 1751.33
Fig 2.2: Sample Acceptance Region
Since sample chi-square lies outside the acceptance region [32, 74] we
reject the null hypothesis i.e.: the ranking of websites in India and the
World does not have any correlation with each other.
PhD. Thesis by Devendera Agarwal 35
To find out, whether e-commerce is really useful other than
convenience, time saving, etc., we visited all the B2C websites and
found various similar products under different categories and also
found their respective prices in surrounding area of living, making
sure that we get the cheapest price and is also convenient to go and
buy. To our surprise e-commerce also turned out to be the cheapest in
all cases, with 7 days replacement guarantee and in most cases free
postage and handling. The results of our finding are shown in Table
2.5 and Figure 2.3 represents the regression line showing relationship
between Category & Prices.
E-commerce Websites (Price in Rs.) S.No Category Model No Rediff
India eBayIndia
India times
SifyLucknowMarket
1 Mobile Phones Nokia E-63 7925 8899 9711 9099 106702 Laptops Compaq 610 32887 30888 33100 34999 334603 Camera Canon
Powershot A3100
6245 8267 8095 6590 8320
4 Watches FastTrack Essential 1230Sl01
1895 2240 2195 2200 2250
(Rates taken in first week of February, 2011, and may change) Table-2.5: Comparative Prices of Various Products
Offered by E-commerce Websites
PhD. Thesis by Devendera Agarwal 36
05000
10000150002000025000300003500040000
Rediff India eBay India Indiatimes Sify Market
Lucknow
Mobile Phones Nokia E-63 Laptops Compaq 610
Camera Canon Powershot Camera A3100
Watches FastTrack Essential Watches 1230Sl01
Fig 2.3: Regression Line Showing Relationship between
Category & Prices
Two other websites of Indian Railway Catering & Tourism Corporation
(IRCTC) and Makemytrip both are related to travel industry i.e booking
of air and rail tickets and reservation in hotels. According to news
nearly 40% of booked tickets are sold online. According to a report by
IAMI (Internet and Mobile Association in India) 75% of total e-
commerce business comes from travel industry and rest everything in
B2C and C2C category contribute only 25%. Indian Railways fourth
largest in the world carries 20 million passengers daily and IRCTC
being official website for booking tickets is no surprise biggest
contributor to e-commerce in India.
In India railways tickets can either be booked at railway station or
online/through agents. Most stations offer booking service from 8 AM
to 8 PM and tickets sold are cheaper then purchased online. But the
shear amount of time it takes to get ticket booked tells upon the
patience of any person at times it takes half/ full day to get tickets
booked thus justifying getting the tickets booked online.
PhD. Thesis by Devendera Agarwal 37
2.4 Conclusion
We have seen that potential for growth of e-commerce in India is
enormous. We have also seen that amount of interest that is there for
travel industry is not seen in other services. Professional e-commerce
websites are doing excellent job but what are the factors that are
inhibiting users from purchasing online need to be ascertained.
***
PhD. Thesis by Devendera Agarwal 38
CHAPTER – 3
TRUST IN E-COMMERCE
Chapter – 3
TRUST IN E-COMMERCE
3.0 Introduction
Trust is related to uncertainty in e-commerce environment. Every new
technology has few pros and cons; internet is also part of this new
scientific innovation that has revolutionized our life. Business houses
are largely relying on this new technology. Internet has become vital
part of our lives these days. But there exists darker side to this
innovation; the use of e-commerce has induced new set of electronic
threats [41, 78]. These threats include misuse of personal data (e.g.
credit card number), web spoofing and purposeful misinformation.
This has led to building of distrust [22, 35] with the e-commerce
framework. A certified server authenticates the server not the web
content information on the server. An authentic server portrays web
user that it is truly the server that it is claiming to be. Similarly the
certificate shown by the website does not promise the authenticity of
product information it displays, it is only concerned about transaction
security in real time mode. The metadata present in Trust Modeling
[77] represent the trustworthiness of a web page and Trust Attributes.
In this chapter we first focus on trust as defined by various authors
and later develop a Fuzzy Regression model [15, 25] of various Trust
Attributes. We will be highlighting the work of Yang, Brown and Lewis
PhD. Thesis by Devendera Agarwal 39
[127] in identifying the Trust Attributes. We also develop our model
TrFRA (Trust Based Fuzzy Regression Analysis).
3.1 Defining Trust
Trust, is a very vital aspect of human life. We use it every day in our
day to day life. For instance, when we go on holidays and trust
transport not to crash, our houses in the mean time remains safe, we
trust our bank not to cheat and some even trust their vehicles to start
every time they start them. Trust is crucial to all transactions, while
our own actions are reliant on the actions of others. Thus, leaving out
cases where trust has no influence on our result. Trust can be strong
or weak depending on the situation. Morton Deutsch defines trust as
follows [64]:
“
If an individual is confronted with an ambiguous path, a path
that can lead to an event perceived to be beneficial (Va+) or to an
event perceived to be harmful (Va-);
He perceives that the occurrence of Va+ or Va- is contingent on
the behavior of another person; and
He perceives that strength of Va- to be greater than that strength
of Va+.
”
In this definition, Deutsch portray an incident with a beneficial
outcome as Va+, while with a harmful result is considered as Va-.
Although this definition is based on psychology but it outlines one of
PhD. Thesis by Devendera Agarwal 40
the most crucial necessity to establish trust. If the visible benefit is
greater than the visible harm then the impact of trust in the selection
would not be that big. In other words, a trust linking requires a
harmful path with more worth than the beneficial one. For example,
an employee would not risk hacking a computer and would rather
switch identity with another person if his salary is less than his own.
A more distinct definition is given by Diego Gambetta [26]:
“Trust (or, symmetrically, distrust) is a particular level of the
subjective probability with which an agent assesses that another
agent or group of agents will perform a particular action, both
before he can monitor such action (or independently or his
capacity ever be able to monitor it) and in a context in which it
affects his own action.
When we say we trust someone or that someone is trustworthy,
we implicitly mean that the probability that he will perform an
action that is beneficial or at least not detrimental to us is high
enough for us to consider engaging in some form of cooperation
with him.
Correspondingly when we say that someone is untrustworthy,
we imply that that probability is low enough for us to refrain from
doing so.”
PhD. Thesis by Devendera Agarwal 41
Mcknight and Chervany [81] as motivation Josang. [63, 64] define
trust as:
“Trust is the extent to which one party is willing to depend on
somebody, or something, in a given situation with a feeling of
relative security, even though negative consequences are
possible.”
This definition might not be as clear but it consists of all the relevant
components of a trust communication. Other than the above
definitions and restrictions, trust is taken care of differently in some
other fields.
By saying “we trust someone” or that “he is trustworthy”, we without
reservation mean that probability of his being performing a task that
is constructive or at least not destructive to us is high enough for us
to consider him associating with us in some form of cooperation.
Similarly by saying that “someone is untrustworthy”, we mean that
that probability of him being associating with us is extremely low.
In a network driven virtual world, where interaction corresponds to
the exchange of messages between digital entities, trust is a basic
challenge; user identity, exchange medium everything is under
suspicion, and with no chance of communication or probability of
future communication between individual entities. Nevertheless, trust
stays a prerequisite for maintaining supportive social groups and
online financial transactions.
The communication between server and client are not secure unless it
is providing a safe and secure transaction. To reduce the risk we must
PhD. Thesis by Devendera Agarwal 42
deal with development of trustworthiness of the web services, which
finally means increasing the complexity of the website.
3.2 Trust in E-Commerce Trust is considered imperative and a main concern within an e-
business because the customers do not interact with the business or
the staff face-to-face. In e-commerce one must create trustworthy
relationships, attract and maintain a customer base within e-
businesses due to the absence of the physical product, where the
customer and seller are physically separated. This can result in an
insecure environment where trust is of high importance. Companies
need to create relationships which result in trust relationships where
initial sales will be generated, which can lead to customer loyalty.
Trust is frequently associated with security worries in online
transactions. A transmission is secure when that two parties
participating in a transaction have been properly verified and that the
information transfer taking place via the network remains unaffected.
However, three main ways in which confidential information can be
obtained [14]:
1. Information copied during transmission
2. Information accessed during storage
3. Information acquired from an authorized party
Whom can we trust? In fact, almost 95% of all security breach
incidents are caused by an insider [10]. It means that properly set
PhD. Thesis by Devendera Agarwal 43
secured systems are still at risk from people with lawful access to the
system.
A major hurdle in full operation of business-to-consumer (B2C) e-
commerce is the lack of trust on consumer end. Trust is developed in
a business through people by their own previous experience with that
business, through reports about that business from trusted third
parties or from experience of other consumers. While a lot of energy
has been spent on “privacy” and “secure transactions”, resulting in
“seals of approvals” and “trustmarks”, lot need to be done in providing
assurance to the consumer that the site they are going to deal is
“legitimate” in various issues such as delivery, return policies, etc.
There are various reasons for the slow growth in business-to-
consumer than in business-to-business e-commerce, but most
important being much slower development in e-commerce supporting
services such as security and payment services on which business-to-
consumer e-commerce depends. These supporting services are
essential in crafting the “legitimacy” conditions mandatory for trust to
develop on the consumer side. Some important legitimacy conditions
[66] that allow trust to develop include:
the sellers are who they state to be
the seller has right of sale over the item in question
the transaction and payment methods available are legal and
secure
information about the buyer is kept confidential and is not sold
to other parties and used only for its intended purpose
PhD. Thesis by Devendera Agarwal 44
the item sold match up to its description and is working and
appropriate for its intended purpose
the purchased item is delivered to the buyer within specified
time frame
the buyers are who they claim to be
the buyer has the funds to purchase the item particularly in
case of COD (Cash on Delivery)
All of above, except last two points are issues of “trust” from the angle
of the consumer, while the last two points are concern of trust from
the angle of the seller. In this research, we have concentrated on the
issue of trust from the consumer perspective.
3.3 Current Approaches for Evaluating Trust Current work on trust issues is highly segmented, with individual
groups concerned more about their own particular fields of inquiry.
Few of these methodologies are discussed below.
3.3.1 Self-Regulation
On-line trade communities can self-regulate, and thereby
encourage the growth of trust among consumers, by
participating in reputation-based schemes that provide a seal
trustmark, once a site has satisfied minimum trust criteria. The
seal is withdrawn if there are any violations.
The W3 Consortium’s “Platform for Privacy Preferences Project”
(P3P) tries to offer a framework for informed online
communication and present means for a website to encode its
PhD. Thesis by Devendera Agarwal 45
data collection and data use practices in a consistent, machine
readable XML format.
The two most admired trust label programs are “TRUSTe”
(www.truste.org) and “BBBOnLine” (www.bbbonline.org).
TRUSTe is a nonprofit organization with mission to construct
user trust in Internet by endorsing the principle of disclosure.
In the TRUSTe system, privacy is the main trust criterion. After
receiving an online company website request, TRUSTe audits
that website. If website adheres to established privacy
principles, i.e. meets the core criteria, and is prepared to abide
by with controlling and consumer resolution procedures, then
the TRUSTe seal will be awarded to the website.
Similarly, the BBBOnLine offers a “privacy seal” to a websites
that placd online privacy policies and adheres to the principles
of the Better Business Bureau i.e Disclosure, Choice and
Security. It also scrutinizes compliance and request specific
sanctions for disobedience.
Trust label programs require vigilance in their monitoring to
ensure that privacy standards are upheld. However recent
surveys have illustrated that people does not seem to appreciate
privacy seal programs.
A negative reputation system [91] has been proposed in which
information on untrustworthy traders is publicly distributed. In
other words, reputation serves as a double edge sword on one
hand its serves as source of information while on other hand it
PhD. Thesis by Devendera Agarwal 46
also serves as a potential source of sanctions. These systems
accumulate, disseminate, and aggregate feedback about buyers’
and sellers’ previous behavior. Hardly any of these people know
each other; this system facilitates them to decide whom to trust,
thereby encouraging trustworthy behavior, and deterring
dishonest partaker.
From a social perspective, Friedman et al [8] believe ‘people trust
people, not technology’. They also suggest online trust can be
cultivated through 10 trust-related characteristics of online
interaction, such as
1. reliability and security of the technology;
2. knowing people online behavior;
3. ambiguous language and images;
4. disagreement about what counts as harm;
5. informed consent;
6. anonymity;
7. accountability;
8. Importance of promptness in the online environment;
9. insurance and accomplishment history and
10. reputation
They point out that we are susceptible to trust violations in two
ways: “loss of money” and “loss of privacy”.
3.3.2 Presentation of Website
Several empirical works on the importance of online trust
factors for keeping customer loyalty [2, 31, 35, 38, 72, 92, 113,
PhD. Thesis by Devendera Agarwal 47
116] have identified a number of customer-interface design and
trust factors, including assurances, references, certifications,
privacy provisions, consumer protection, and security policies.
Head et al, have an understanding of the way trust is built
through humanized website design [65, 86]. They believe there
is some relationship between human elements in user interface
design. Online trust and User Confidence can be built through a
humanized website featuring several factors: brand, assurances,
fulfillment and website design.
3.3.3 Agent-Based User TechnologyAgent technology researchers [6, 24, 29] have tried to develop
‘smart and intelligent’ agents that could be combined with other
technology such as “virtual reality” to facilitate trust based on
“promises being made, enabled and fulfilled” to build a trusted
customer relationship between customers and providers. This
research considers the foundation stone for a successful and
long association with the consumer is trust, as it could
establish the consumer future behavior and devotion towards
the business.
Although the work reveals some interesting ways of
implementing Web trust through agent technology, Web users
may have difficulty accepting this kind of external manifestation
of trust agent, which does not address the issue of why
technologies should be trusted. In addition, replacing web-
based interfacing with an agent-based interface may not be
PhD. Thesis by Devendera Agarwal 48
feasible for a considerable time because of inertia between the
implementation of a particular approach and acceptance by
general Web users. However, over time this mismatch could be
overcome as a result of increased knowledge by Web users.
3.3.4 Mathematical ApproachesA number of trust models including Josang 1999 and Abdul
Rahman & Hailes mainly address the problem of an entity’s
identity by using cryptographic mechanisms for propagating
trust measures that take place within the information security
community.
Josang’s “beliefs” model is based on subjective logic, an
extension of standard logic and to a certain degree, probability
theory. The model may prove suitable for assigning trust values
when we are dealing with uncertainty. The proposed back-
propagation [84] method seeks to automatically produce a
metadata description, thus making it easier to classify Web
information by “fuzzifying” the metadata attributes.
3.3.5 Public Key Infrastructure and SecurityVarious trust models use public key encryption to present a
Web authentication framework, and tries to attain the
maximum trust with minimum risk. These trust models include
the “X.509 standard Public Key Infrastructure” (PKI) trust model,
the “Pretty Good Privacy” (PGP) trust model, the “Simple Public
Key Infrastructure” (SPKI) and the “Simple Distributed Secure
Infrastructure” (SDSI). However we must state that PKI trust
PhD. Thesis by Devendera Agarwal 49
models could not achieve 100 per cent trust. Gutmann [42]
stated,
“CA certificates can exhibit numerous vulnerabilities.
Because all CAs are assigned the same level of trust, the
entire system is only as secure as the least secure CA”.
Nonetheless, there is a requirement of variety of PKI trust
models, whether it is formal hierarchy used in X.509 or the Web
of trust used by PGP or a simple data structure used by SDSI &
SPKI. These PKI trust models offer dissimilar structures of trust
and take diverse approaches towards creating a trust
relationship. Their certification services also offer great amount
of variety of hierarchy of trust between business parties on the
Web.
Many researchers propose various approaches to address different
trust issues in the eCommerce environment, including a better
human-oriented front-end design, improved technologies for online
transaction, and public user protection policies. However, the
trustworthiness of webcontent cannot be addressed by online security
technologies alone.
3.4 Fuzzy Regression To arrive at a solution for this problem we concentrated on Fuzzy
Logic which provides a simple yet definitive way to arrive at a
conclusion based upon vague, ambiguous, imprecise, noisy, or
PhD. Thesis by Devendera Agarwal 50
missing input information. This approach to control problems
impersonates how a person would make decisions, only much faster.
Regression Analysis is a statistical technique used for estimating the
relationships among various variables. It includes various techniques
for modeling and analyzing several variables, with the focus on the
relationship between a dependent variable and one or
more independent variables. Regression analysis helps us to
understand how the typical value of the dependent variable changes
when any one of the independent variables is varied, while the other
independent variables are held fixed. While Regression Analysis can
only fit crisp data but Fuzzy Logic can be used to fit both fuzzy data
and crisp data into a regression model thus solving our problem.
In Regression Analysis the dependent variable ‘Y’ is the function of the
independent variable; the degree of contribution of each variable to
the output is represented by coefficients on these variables. Normal
regression model is shown as
nnxaxaxaaaxfY ..),( 22110 (3.1)
The conventional Regression Analysis is probabilistic. In Fuzzy
Regression, the difference between observed and the estimated values
is assumed due to ambiguity present in the system. The fuzzy linear
regression model in represented as:
0 1 1 2 2~ ~ ~ ~~ ~( , ) ... n nY f x A A A x A x A x
(3.2)
Fuzzy regression estimates [99] arrange of possible values that are
represented by a possibility distribution. Membership functions are
PhD. Thesis by Devendera Agarwal 51
formed by assigning a specific membership value to each of the
estimated values. In this analysis we will be analyzing the Triangular
Membership Function.
The membership function ~A for each coefficient is expressed as:
~( )iA a| |1 ,
0
i ii i i i i
i
p a p c x p cc
Otherwise (3.3)
The above formula is derived from the fuzzy function which consists
of two parameters ‘p’ and ‘c’. The spread of the triangle in fig. 3.1
denotes the fuzziness of the function.
~A
1.0
Fig. 3.1: Membership function of fuzzy coefficient ~A
The fuzzy parameters 1~ ~ ~
{ , }nA A A can be denoted in the vector form of
~ { , }A p c where 0{ ...... }np p p and 0{ ...... }nc c c . The updated
equation of is given as: ~Y
0 0 1 1 1 2 2 2~ ( , ) ( , ) ( , ) ..( , )n n nY p c p c x p c x p c x (3.4)
ci pi ci
O a
~iA
~A
PhD. Thesis by Devendera Agarwal52
The membership function for the output fuzzy parameter [73] is
given as:
~Y
~~m ax(m in ( )
( )0
i iiA a
Y y
(3.5)
The purpose of the fuzzy linear regression model [36, 62] is to
ascertain fuzzy parameters *
~A that minimize the spread subject to the
constraint.
3.4 Trust Attributes For analysis of Trust and its impact on Indian user we use three
categories.
3.4.1 Web Object Content (WOC) It deals with knowledge derived about the web document that can be
divided into different categories like when it is created as frequent
information change is an essential requirement these days.
3.4.2 Web Object Ownership (WOO) This attribute focuses on reputation of the firm and its type, for
example, .gov, .com, .edu, .org etc. In India the ‘.gov’ and ‘.edu’ are
considered safe as they generally do not have false information on
their website.
3.4.3 Web Object Certification Authority (WOCA) This is one of most important attribute of a Trust Model. Here main
focus is on certificate, as a user main task is to give special care to
type of certification the server is having and the certification of the
web document.
PhD. Thesis by Devendera Agarwal 53
3.5 TrFRA Model Let w1, w2 and w3 denote the Trust features of Web Object
Content, Web Object Ownership and Web Object Certification
Authority respectively. Since the data sets gathered from the users
are non-fuzzy in nature we will use Tanaka’s Regression Model.
3.5.1 Brief Overview of Tanaka Model For a non fuzzy data the objective of the regression model is to
determine the optimum parameters *
~A such that the fuzzy output set,
which contains ‘yi’ is greater then ‘h’ where ‘h’ is given by the relation
as shown in fig. 3.2:
1.0
0.5
O
h
µ
A
1
n
i iji
c x1
n
i iji
c x
yj
Y
Fig. 3.2: Fuzzy Output Function
The degree of ‘h’ is specified by the user as ‘h’ increases, the
fuzziness of the output increases, Equation-6 states the fuzzy output
should lie between A and B. The middle value is given as and
the spread as .
1
n
i ii
p x
1| |
n
i iji
c x
~( ) , 1,2,...iY y h j m(3.6)
PhD. Thesis by Devendera Agarwal54
The objective function that needs to be minimized is given as:
1 0min
m n
o ij i
O mc c ijx
1i ij
ix
1
(3.7)
The objective function can be minimized subject to two constraints
given as:
1(1 )
n n
j i iji
y p x h c (3.8)
1(1 )
n n
j i ij i iji i
y p x h c x (3.9)
Since each data set produces two constraints, there are a total of 2m
constraints for each data set.
3.5.2 Case of Lucknow City with Trust Rating on eBayIn order to apply TrFRA Model we have done a sample survey varying
from heavy online shoppers to moderate online shoppers. Based on
the survey we found out that Trust variable focused more on only two
variables i.e. Web Object Ownership (WOO) and Web Object
Certification Authority (WOCA). We get five data samples as shown in
Table 3.1.
PhD. Thesis by Devendera Agarwal 55
T WOC WOCA
25 3.5 5.0
30 6.7 4.2
11 1.5 8.5
22 0.3 1.4
27 4.6 3.6
19 2.0 1.3
Table 3.1: Fuzzy Regression Value of Sample Data
In this case there are five data sets hence the number of constraints
will be 2x5=10 constraints. The fuzzy linear regression [79, 118]
equation-2 is used to fit the data set.
Solving with the Fuzzy Logic Tool Set from MATLAB we get the values
WOC & WOCA from which, we are able to derive the equation-3.10 for
trust as:
~ ~20.3916 2.340 1.328T WOC ~WOCA (3.10)
3.6 Conclusion This chapter focuses on TrFRA which gives a relationship model in
terms of Fuzzy output and regression line. The output represents
TRUST factor and its correlation with two factors WOC and WOCA
also the estimated value for the output variable. TRUST is derived by
the variation between the observed and the estimated value. It is
PhD. Thesis by Devendera Agarwal 56
assumed to be due to the ambiguity intrinsically present in the
system.
The output TRUST for a specified input is assumed to be a range of
possible values i.e. output can take on any of the possible values.
The advantage with Fuzzy Regression is the range of Possibilistic
values is much more as compared to the Normal Regression Model.
***
PhD. Thesis by Devendera Agarwal 57
CHAPTER 4
TRUST V/S COMPLEXITY OF
E-COMMERCE SITES
Chapter – 4
TRUST V/S COMPLEXITY OF E-COMMERCE
SITES
4.0 Introduction
India today is facing with various kinds of threat to e-commerce
systems. The problem arises when we increase the security of the e-
commerce website, the complexity at the user level also increases,
which in turn affects the volume of sale. While traditional marketing
does not involve any type of complexity since the consumer deals
directly with the supplier. Since internet marketing does not involve
any face to face direct interaction so a visual interface is essential.
There are various types of online buying behavior models like Bettman
(1979) and Booms (1981) in which the focus was on personal
characteristics viz. Culture, Social Group and Physiological Behavior.
Lewis and Lewis (1997) have classified web users in five categories:
Directed information-seekers: It includes those users who are
looking for product information only and normally not planning
to buy online.
Undirected information-seekers: Such users commonly
referred as 'surfers', generally ends up on a particular website
accidentally by browsing and following hyperlinks. Users of this
group tend to be inexperienced users and also likely to click
banners of the website.
PhD. Thesis by Devendera Agarwal 58
Directed buyers: These are actual buyers, who are online to
purchase specific products. Such users would visit location that
compares product features and prices.
Bargain hunters: These users love to hunt the offers available
from sales promotions such as free samples or contests.
Entertainment seekers: Such users interact with web for
seeking fun through entering contests such as quiz’s, puzzles or
interactive games.
Under all the above categories the main focus is the trust of web users
which will finally lead to purchase.
The communication between server and client are not secure unless it
is providing a safe and secure transaction. To reduce the risk we have
to deal with development of trustworthiness of the web services, which
finally means increasing the complexity of the website.
LOW HIGH
HIGH
LOW
If TRUST level is HIGH and COMPLEXITY is LOW consumer will PREFER this.
If TRUST level is HIGH and COMPLEXITY is HIGH consumer will RESIGN.
If TRUST level is LOW then COMPLEXITY has to be LOW consumer will AVOID this.
If TRUST level is LOW and COMPLEXITY is HIGH consumer will ABANDON.
Fig. 4.1. Trust/ Complexity Matrix.
PhD. Thesis by Devendera Agarwal 59
From the figure 4.1 we can conclude that there has to be some
situation in which a trade off between Trust Level and Complexity of
the transaction has to be maintained. This trade off can be achieved
by the help of development of Fuzzy Rule base, but simple Fuzzy Rule
base will not be sufficient for this purpose, so we extend this problem
and solve it using Evolutionary Multi-objective Optimism [45, 83, 98].
4.1 Trust Issues in Ecommerce Web trust issues include user confidence, privacy, online security,
and authenticity of business partners, service providers and products.
Trustworthiness of the webcontents could be viewed as a foundation
of Web trust. Web trust can be described as a counterbalance to
elements of uncertainty. All Web communities have overwhelmingly
agreed that “eCommerce is a matter of trust”.
The Web is becoming increasingly important in providing the
infrastructure for electronic commerce. The Web offers an unmatched
business opportunity for small and medium-size businesses, and
more and more of them are taking advantage of the low cost of
establishing a Web-based business.
Electronic Commerce (eCommerce) has changed the traditional service
relationship between consumers and service providers and
consequently, the conventional way of assessing trust in a business
relationship. For example, the traditional way of verifying legitimate
business identities was based on the physical uniqueness, such as the
shop front of a business provider. On the Web, business providers and
PhD. Thesis by Devendera Agarwal 60
consumers may no longer be able to be recognized by conventional
means (i.e. their physical appearance); instead, their identity is
apparent in their websites, email addresses or by some electronic
means (e.g. an electronic token, commonly a public key or digital ID).
These changes have brought a range of potential risks to Internet
users, including fraud, misuse of personal data (e.g. credit card
numbers), deliberate misinformation, Web spoofing (i.e. copying
legitimate businesses in order to unlawfully obtain credit card
numbers), eavesdropping, identity theft, repudiation, unlawful
webcontent modification, masquerading and insecure transmission.
These perceived threats are partly the result of a lack of rules, online
service standards, codes of conduct (or protocols) and the ability to
police them, creating ‘uncertainty’ in the eCommerce environment.
Some of the elements of this uncertainty are easily identified, while
some are hidden or embedded in the eCommerce environment (e.g.
misinformation hidden in the webcontent). This uncertainty affects
consumer confidence in the online ecommerce.
Trust requirements can be based on various consumer expectations
and user-confidence improvement factors. A national survey of
Internet users for consumers by Webwatch identified nine deciding
factors that Web users rely on when visiting a website:
ease of navigation;
have trust on information displayed on a website;
is able to spot the sources of information on a website;
know that the website is maintained & updated regularly;
PhD. Thesis by Devendera Agarwal 61
is able to locate the important information about a website;
have knowledge about the ownership of website;
have knowledge of bankers and organizations financially
supporting the website;
the presence of trust seals from trusted parties; and
the presence of awards and accolades from trusted parties.
The above surveys provide empirical evidence of Web users’
perspectives on online trust and trustworthiness requirements, for the
benefit of both web users and providers, i.e. for service providers, a
better user-interface design; for Web users, a basic judgment of the
trust worthiness of a website.
However following tasks are important in building the legitimacy
conditions needed for trust to develop on the consumer part. Few
significant legitimacy conditions that permit trust to develop consist
of:
the sellers are who they state to be
the seller has right of sale over the item in question
the transaction and payment methods available are legal and
secure
information about the buyer is kept confidential and is not sold
to other parties and used only for its intended purpose
the item sold match up to its description and is working and
appropriate for its intended purpose
PhD. Thesis by Devendera Agarwal 62
the purchased item is delivered to the buyer within specified
time frame
the buyers are who they claim to be
the buyer has the funds to purchase the item particularly in
case of COD (Cash on Delivery)
All except last two items are subject of “trust” from the point of the
consumer, while the last two points are concern of trust from the
seller point of view. In this research, our focus is on the issue of trust
from the consumer perspective.
4.2 Existing Web Based Trust Models Many trust-related projects involve researchers, practitioners,
industries and governments. The World Wide Web continues its efforts
in every possible Web-related field to advance Web technology. The
following projects have explored various ways to address Web trust.
The REFEREE Project (REFEREE: Rule-controlled
Environment for Evaluation of Rules, and Everything Else) [119]
by W3C working groups. It provides a general mechanism for
expressing and assessing trust management policies. The
mechanism may be used to solve problems related to trust
management that exist in the World Wide Web by building a
trust infrastructure i.e. protocols, policy, languages, execution
environment and metadata format for all Web applications
requiring trust. REFEREE checking user policies in response to
host application’s request for action. Policies are generally
regarded as programs in REFEREE. For any given request,
PhD. Thesis by Devendera Agarwal 63
REFEREE calls upon the suitable user policy and interpreter
module that returns the answers host application question of
whether or not the request fulfill with the policy.
The DSig project [106]: The DSig project uses digitally signed
labels to make authenticate able declaration about unrelated
documents or about patent of aggregate objects.
PICS [90] was developed by the W3 Consortium [W3C96] to
address the problems pertaining to protecting children from
pornographic material on the Internet without violating the right
to freedom of speech.
The PolicyMaker [13] which is based on PICS and was initially
designed to deal with trust related problems in network services
that process signed requests for action and use public-key
cryptography. It specifies what a public key is certified to do
[12].
KeyNote [11], the successor to PolicyMaker, was developed by
AT&T Research laboratories to work upon the weaknesses of
PolicyMaker with two additional design goals: “standardization”
and “ease of integration”.
Recreational Software Advisory Council: This model is used
by parents and teachers for content filtering on the Web. In
1999 it was merged into a new organization, the “Internet
Content Rating Association” (ICRA). The original aims of RSAC,
to guard children from possibly harmful data while safeguarding
free speech on the Internet, continue to provide the cornerstone
PhD. Thesis by Devendera Agarwal 64
for ICRA’s work. The RSACi system (RSAC on the Internet) has
been incorporated into Netscape Navigator and Microsoft’s
Internet Explorer, the latter since the release of version 3.0 in
February 1996.
Much research has focused on the area of consumer confidence and
behaviors. From the Web consumer’s viewpoint, likely concerns could
be related to privacy, security, authenticity of the service providers,
and trustworthiness of published information on websites.
4.3 Our Proposed Model Genetic Algorithms [27, 88, 125] have been frequently used to model a
solution for conflicting goals. Genetic Algorithms uses experienced
based techniques for problem solving and try to emulate how an
experienced person would have behaved under the same situation.
Let Trust (T) be a measure of security which the customer will be
provided and Inverse of Complexity (C) be the user comfort level.
Applying the Fuzzy Rule base we can get
Maximize Trust (T) (4.1)
But it leads to compromise in the complexity (C) of fuzzy rule based
systems [60, 61, 109, 125]. According to consumers survey most of
consumers in India considers Trust and Ease of Use (Lower level of
Complexity) at the same time. The above problem can be formulated
as
Maximize Trust (T) subject to
Inverse complexity (C) (4.2)
PhD. Thesis by Devendera Agarwal 65
where complexity (C) is the measure of fuzzy rule system.
We can develop a single objective function to the above solution given
as:
Maximize ƒ(Trust (T), Inverse of Complexity (C)) (4.3)
We can also use weights in order to determine the exact function for e-
commerce site. Let w be the weight,
Maximize (w1) Trust (T) +
(w2) Inverse of Complexity (C) (4.4)
We proceed with development of more refined stages in which we can
focus on various stages of membership functions. Consider a simple
single output function y = f(x) an application of Takagi-Sugeno
method [81, 109, 115, 125] we can write it as:
Rule Ri : if x is Ai then y=ai+bjx, i=1,2,…N
Rule Rk : if x is Ak then y=ak+bkx, k=1,2,…N
:
:
:
:
Rule Rz : if x is Az then y=az+bzx, z=1,2,…N (4.5)
This output value is given as:
N
iAi
Ai
N
ixi
x
xbaxy
1
1
)(
)()()(
(4.6)
PhD. Thesis by Devendera Agarwal 66
where y(x) is the estimated output value for the input value x
and μAi (x) is the membership value of the antecedent fuzzy set
Ai.
From the input-output data we can derive the relationship between
Trust and Complexity of the e-commerce site considering three
Takagi-Sugeno Rules.
We develop a heuristic rule [107, 114] denoted by three lines A, B and
C as the subsequent of the linear function with fuzzy sets A1, A2 and
A3. Each of the Fuzzy Rule can be represented in triangular Fuzzy
Sets.
Rule R1: If TRUST is SMALL and COMPLEXITY is HIGH Then
User’s Ease of Use is LOW.
Rule R2: If TRUST is LARGE and COMPLEXITY is SMALL Then
Users Ease of Use is HIGH.
Rule R3: If TRUST is SMALL and COMPLEXITY is SMALL Then
Users Ease of Use is HIGH.
A B C
Fig. 4.2. Three Takagi-Sugeno Rules
PhD. Thesis by Devendera Agarwal 67
Based on the above rules we try to develop a plot between Complexity
and Trust and develop our interpretable solution [43, 69, 82, 124,
126] between the two entities.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Trust
Com
plex
ity
Fig.4.3. Input Output Data using Fuzzy Data Set
Possibly we can also merge the above set of rules to achieve more
refined results, but a relationship generated by optimization rules
gives some gridlines in the area of relationship between the two
entities.
4.4 Conclusion This chapter highlights the importance of Trust on any e-commerce
website. We also tried to derive the relationship between Trust and
Complexity and its effect on consumers. It is very difficult to interpret
the exact relationship between the two entities. Different Fuzzy rule
PhD. Thesis by Devendera Agarwal 68
are being applied in order to determine the appropriate
interpretability. The method that we have used is the application of
Fuzzy Optimization Theory [58, 59, 68, 80] to find the probable
relationship between Complexity and Trust. The future extension
would be to use Evolutionary Algorithm [61, 70] in finding out the best
possible trade-off between the two entities.
***
PhD. Thesis by Devendera Agarwal 69
CHAPTER 5
WEB PERSONLIZATION AND
RECOMMENDATION
Chapter – 5
WEB PERSONILIZATION AND
RECOMMENDATION
5.0 Introduction
Electronic Commerce is fast emerging as most popular method of
purchasing, let it be a small pen drive or bulky LED TV. Recent survey
[55] has estimated that around 3-5% of Indians have transacted or are
well versed with working of online shopping websites. The strategy
which is being followed until now related to the various policy
initiatives like:
Consumer Proportion: This model is being propagated by the
government based on certain guidelines for the protection of
consumers.
Legality: It deals with formal recognition of electronic
signatures; In India digital signatures are necessary for e-
Tendering.
Security: Central Government has issued its policy relating to
cryptography techniques to ensure secure electronic commerce
in third party transfer.
In order to deal with security and web personalization [17] issues we
develop two basic classification methods: Naïve Bayes and K-nearest
neighbor. We start this chapter with the introduction to Web
Personalization and Recommendation System. Later in this chapter we
PhD. Thesis by Devendera Agarwal 70
highlight the classification methodologies using Bayesian Rule for
Indian e-commerce websites. Firstly, it deals with generating cluster of
users having fraudulent intentions. Secondly, it also focuses on
Bayesian Ontology Requirement for efficient Possibilistic Outcomes.
5.1 Web Personalization & Recommendation System
Web personalization, consisted of activities such as providing
customized information, changing the webpage layouts and adapting
the contents tailored to the user’s need, has become an essential part
of a website to enhance its compatibility and attractiveness. The
recommendation system [108] or interactive decision aids [46] can be
considered as one form of personalization to facilitate in helping the
users making purchase decisions.
Main differences between the traditional brick-and-mortar stores and
e-commerce websites are the infinite shelf-space on the Web. Unlike
the traditional stores which have limited storage, the E-commerce
websites provide the consumers a wide variety of options, alternatives
and product information. The diversity of product choices and the
abundance of messages on an e-commerce site have led to the
problem of overloading. To overcome this problem demand of web
personalization and real-time adaptation catering to the user’s need
has arise. Reason being shopping experience can be overwhelming
especially when there is no assistance available in deciding what
products to purchase. In addition, the effort and time spent on
searching aimlessly may lead to poor quality of decision and
PhD. Thesis by Devendera Agarwal 71
dissatisfaction of the consumers [17]. Therefore, to find the ideal
products in mind effectively and efficiently, online customers not only
look for the suggestions from their peers, and editorial picks [112] but
also heavily count on the real-time recommendation systems featured
on the e-commerce websites [46, 47, 57, 100, 111].
Recent advancement in web technology helps online companies to
obtain individual customer’s information in real time. Based on
acquired information, they build detailed profiles and offer
personalized services. Thus e-shops now have the prospect to increase
their performance by focusing on individual user preferences and
needs, thus increasing satisfaction, improve loyalty, and creating one-
to-one relationships.
A recommendation system is a system or application that helps the
user to select a suitable item or finding relevant information among a
set of candidates using a knowledge-base that can either be hand
coded by experts or learned from behaviors of the users. Typically, a
recommendation system performs three of functions:
Information Collection: The recommendation system collects
all the usable information for the prediction task including the
users’ attributes, behaviors, or the content of the resources the
user accesses.
Learning: It applies a learning algorithm to filter and exploit the
users’ features from the collected information.
Prediction: It implies the kind of resources the user may prefer
are then made either directly based on the dataset collected in
PhD. Thesis by Devendera Agarwal 72
the information collection phase (memory-based predictions) or
with a model learned from it (model-based predictions).
Recommender systems are generally used by E-commerce sites for
suggesting various products to their consumers and also to provide
them with information to help them decide which products to
purchase. The products suggested can be chosen based on bestsellers
on a site, on the locality of the consumer, or on basis of the previous
buying pattern of the consumer as a prediction for future buying
behavior. Various forms of recommendation consist of suggesting
products to the consumer, presenting tailor made product information
and providing community reviews. Generally, these recommendation
techniques are part of personalization on a website because they help
the website to acclimatize itself to individual customer.
Recommender systems are analogous to, but at the same time unlike
from, marketing systems, supply chain systems and decision support
systems. While marketing systems helps the marketer in making
decisions about how to promote products to consumers, usually by
dividing a broad target market into subsets of consumers who have
common needs and grouping the products in categories that can be
associated with the marketing segments. Later on marketing
promotion can then be run to further encourage consumers in various
segments to purchase products from groups selected by the marketer.
On the contrary, recommender systems directly interact with
consumers, assisting them to choose products they will like to
purchase. Recommender systems generally consist of processes that
PhD. Thesis by Devendera Agarwal 73
are carried manually, such as creating cross-sell lists and procedures
that are performed largely by computer, such as collaborative filtering.
Recommender systems increase E-commerce sales in three ways:
Convincing Browsers into Buyers: Visitors to a Web site often
stay around the site without purchasing anything.
Recommender systems can persuade consumers to find
products they wish to purchase.
Increasing Cross-sell: Recommender systems improve sales by
suggesting extra products for the customer to purchase. If the
recommendations are good, it will increase the average order
size. For example, a site might recommend additional products
based on products already in the shopping cart during checkout
process.
Building Loyalty: In a world where a site’s rivals are only few
clicks away, acquiring consumer loyalties is an essential
business strategy. Recommender systems develop loyalty by
building a value-added relationship between the site and the
consumer. Sites often make investment in learning about their
consumers, use recommender systems to learn consumer
behavior, and to develop powerful custom interfaces that match
consumer needs. Consumers reimburse these sites by visiting
back those that best match their requirements. The more a
consumer involved with the recommendation system – teaching
it what he wants – the more loyal consumer becomes to that
site.
PhD. Thesis by Devendera Agarwal 74
5.1.1 Types of Recommendation System
Types of recommendation system can be classified based upon
techniques used for recommending
5.1.1.1 Content-Based Filtering
The Content-Based Filtering (CBF) makes recommendation
based on the correlation between difference resources. In
content-based recommendation systems, resources are
described as a vector of attributes. The system then learns
profile of the users interests based upon the features presented
in the objects that user has rated. When making a prediction on
the customers’ preferences, the system analyzes the relationship
between the products rated by the users and other products by
calculating the similarity between their attribute vectors. The
type of user profile derived by a content-based recommender
depends on the learning method employed. Decision trees,
neural nets, and vector-based representations have all been
used.
A central problem in content-based recommendation systems is
the need to identify a sufficiently large set of key attributes.
When the set is too small, there is insufficient information to
learn the customer profile. Therefore, content-based
recommendation systems cannot be used for new customers
who purchased only once, potential customers who visit the web
PhD. Thesis by Devendera Agarwal 75
site but have not made any purchase, and customers who want
to buy a product that is not frequently purchased.
5.1.1.2 Collaborative Filtering
The collaborative filtering (CF) is widely utilized and most
established of the information filtering technologies.
Collaborative recommendation systems collect ratings or
recommendations of objects; identify similarities between users
based on their ratings, and produce new recommendations
based on inter user comparisons. A classic user profile in a
collaborative system is made up of vector of items as well as
their ratings, constantly modified as the user interacts with the
system over time.
Collaborative filtering algorithms are categorized into two
classes: “memory-based” and “model-based”. Memory-based
algorithms function over the complete user database to make
predictions. Popular memory-based models are founded on the
concept of nearest neighbors, using selection of distance
measures. Model-based systems are founded on a compact
model deduced from the data, which have used an array of
learning techniques consisting of neural networks, latent
semantic indexing and Bayesian networks.
Collaborative techniques work fine for complex objects such as
movies or music, where dissimilarity in taste is accountable for
much of the change in liking.
PhD. Thesis by Devendera Agarwal 76
The major difference between collaborative and content based
filtering systems is that the collaborative systems follow past
actions of a group of consumers to make a recommendation for
particular person of the group. Using this method, users may
now be able to obtain recommendations for products that are
different in content to those they have previously rated, as long
as other similar minded consumers showed their interests in
these products.
The collaborative filtering recognizes consumers with similar
interests to those of a given consumer, and recommends liked
products by the given customer. However, as most existing
Content Filtering algorithms keenly rely on the user’s ratings on
products to make recommendations, their performances perish
severely when the user rates few items in the database, which is
called cold start problem in the Content Filtering research.
5.1.1.3 Knowledge-Based Recommendation
The knowledge based recommendation tries to propose objects
based upon inferences about users needs and liking. Knowledge
based methods are different in that they have functional
knowledge: “they have knowledge about how a particular item
meets a particular user need”, and can thereby question about
the relationship between a requirement and a possible
recommendation. The user profile can be any knowledge
structure that maintains this inference. In the normal case it
PhD. Thesis by Devendera Agarwal 77
may be a simple query that the user has written while in others,
it may be a more exhaustive representation of the user’s needs.
5.1.1.4 Utility-Based Recommendation
The utility based recommendation does not try to construct long
term picture about their users, but rather use their advices on
an assessment of the match between the user’s needs and the
set of options on hand. The utility based recommendation offer
suggestions by working out the utility of each object to the user.
The advantage of utility based recommendation is that it can
analyze non-product attributes, such as seller reliability and
item availability, into the utility computation, thus making it
possible, to trade off price against delivery schedule for a user
with an immediate need.
5.1.1.5 Demographic Recommendation
The demographic recommendation systems aims to classify the
user based upon personal attributes and then give suggestions
based on demographic classes. The users’ responses are
compared with a collection of manually collected user
stereotypes.
The representation of demographic information in a user model
can vary greatly. Demographic techniques form “people-to-
people” association like collaborative ones, but use different
data. The major advantage of demographic technique is that it
may not need a history of user ratings of the type as required by
collaborative and content-based techniques.
PhD. Thesis by Devendera Agarwal 78
5.2 Our Proposed Model
In order to make our model more illustrative we are taking example of
“Predicting Fraudulent Transaction”.
An Indian e-commerce company has a very large customer base; each
customer has to submit his personal information before making a
transaction. In this way each company is acting as a record and the
response of internet is given as
Z = {Fraudulent, Trustworthy} (5.1)
these are the classification in which we can categorize a customer. By
analyzing from a sample e-commerce site we are able to find out that
in case of Fraudulent the customer-id should be reposted to the e-
fraud cell. Two set of data are taken to check the consistency of data
as depicted in the Table 5.1.
Reporting to
e-fraud cell
No Reporting
RequiredTotal
Fraudulent 20 80 100Trustworthy 100 300 400
Total 120 380 500
Table 5.1: Report of Customer on E-commerce Site
5.3 Naïve Bayes
In order to classify record into ‘m’ classes by ignoring all predictor
information X1, X2,….., Xp is to classify the record as a member of
majority class. For example in our case naïve rule would classify all
PhD. Thesis by Devendera Agarwal 79
the customers to be “Trustworthy”, because 90% of the companies
were found to be Truthful.
Naïve Bayes classifier [9] is an advanced version of Naïve rule. The
logic to introduce Bayes is to integrate the information given in the set
of predictors into the naïve rule to obtain more accurate
classifications. The methodology suggests in finding out the
probability of record belonging to a certain class is evaluated on the
prevalence of that class along with additional information that is being
given on that record in terms of X information.
Since our dataset is very large we prefer Naïve Bayes method. In a
classification task our goal is to estimate the probability of
membership to each class given a certain set of predictor variables.
This type of probability is called a conditional probability. In our
example we are interested in P (Fraudulent | Reporting to e-fraud
cell). In general, for a response of ‘m’ classes C1, C2, ….., Cm and the
predictors X1, X2, ….., Xp we compute as:
P (Ci | X1,…,Xp) where i = 1, 2 , …, m. (5.2)
When the predictors are all categorical we can use a pivot to estimate
the confidential probabilities of class membership. Consider its
application in our example we compute the probabilities divided into
two classes as:
For
P (Fraudulent| Reporting to e-fraudulent cell) = 20/120
and
P (Trustworthy| Reporting to e-fraudulent charges) = 100/120.
PhD. Thesis by Devendera Agarwal 80
The above statement indicates that although the firm is still more
likely to be Trustworthy than Not Trustworthy, the probability of its
being Truthful is much lower than the naïve rule.
However, the method usually gives good result partly because what is
important is not the exact probability estimate but the ranking for
that case in comparison to others.
In order to convert the desired probabilities into class probability we
use Bayes Theorem. The Bayes Theorem gives us the following
formula to compute the probability that the record belongs to class Ci:
)()|,...,(...)()|,...,()()|,...,(
),...,|(1111
1111
mmPP
PPi CPCXXPCPCXXP
CPCXXPXXCP
(5.3)
Ci: To compute the numerator we filter two pieces of information
i) The proportion of each class in the population [P(C1)……P(Cm)]
ii) The probability of occurrence of the predictor vales X1, X2, …,
Xp within each class from the training set.
We develop Table 5.2 of the Users which is categorized as “Frequent
Buyers” and “Occasional Buyers”, for each of these two categories of
Buyers we have information on whether or not reporting has been
done, and whether it turned out to be Fraudulent or Trustworthy.
PhD. Thesis by Devendera Agarwal 81
Reporting to
e-fraud cell User-Type Status
Yes Occasional Buyer Fraudulent
No Occasional Buyer Trustworthy
No Frequent Buyer Fraudulent
No Frequent Buyer Trustworthy
No Occasional Buyer Trustworthy
No Occasional Buyer Trustworthy
No Frequent Buyer Trustworthy
Yes Occasional Buyer Fraudulent
Yes Frequent Buyer Fraudulent
No Frequent Buyer Fraudulent
Table 5.2: Sample of 10 Users for fraudulent intentions
The probability of fraud can be defined by four possible states {Yes,
Occasional Buyer}, {Yes, Frequent Buyer}, {No, Occasional Buyer}
and {No, Frequent Buyer}.
i) P(Fraudulent | Reporting = Yes, Customer Type = Occasional
Buyer) = 1/2 = 0.5
ii) P(Fraudulent | Reporting = Yes, Customer Type = Frequent
Buyer) = 2/2 = 1
iii) P(Fraudulent | Reporting = No, Customer Type = Occasional
Buyer) = 0/3 = 0
iv) P(Fraudulent | Reporting = No, Customer Type = Frequent
Buyer) = 1/3 = 0.33
PhD. Thesis by Devendera Agarwal 82
We can extend this for Naïve Bayes probabilities, for analyzing the
conditional probabilities of fraudulent behavior “Reporting to e-fraud
cell” = Yes, and “User Type” = Occasional Buyer, the numerator is a
proportion of “Reporting to e-fraud cell”. Instances amongst the type
of Buyers, times the proportion of Fraudulent Customers
= (3/4) (1/4) (4/10) = 0.075
To get the actual probability we calculate the numerator for the
conditional probability of truth given
Reporting to e-Fraudulent Cell = Yes;
Type of Customer = Occasional Buyer;
The denominator is then the sum of two conditional probabilities
= (0.075 + 0.067) = 0.14
Therefore the conditional probability of fraudulent behaviors is given
by
PNB (Fraudulent | Reporting to e-Fraudulent cell = Yes;
Buyer Type = Occasional)
= (3/4) (1/4) (4/10) (3/4) (1/4) (4/10) + (1/6) (4/6) (6/10)
= 0.075/0.14
= 0.53
PNB (Fraudulent | Reporting to e-Fraudulent cell = Yes;
Buyer Type = Frequent) = 0.087
PNB (Fraudulent | Reporting to e-Fraudulent cell = Yes;
Buyer Type = Occasional) = 0.031
PhD. Thesis by Devendera Agarwal 83
Rank Ordering of probabilities are found to be even closer to exact
Bayes method than are the probabilities themselves, to further
analyze we can use classification matrix.
5.3.1 Advantages & Disadvantages of Naïve Bayes Classifier
An advantage of the Naive Bayes Classifier is that it only needs a
small amount of data to train and estimate the parameters (Mean and
Variances) of the variables required for classification. Since
independent variables are assumed, only the Variances of the
variables for each class need to be ascertained and not the complete
covariance matrix. The logic of using Naïve Bayes Classification
Technique [121] is to attain computational efficiency and good
performance.
5.3.2 Fuzzy Information Classification and Retrieval Model
The above section deals with a classification technique [105] by which
we can categorize the customer visiting our site based on their
transaction history. The distinctiveness of Fuzzy systems that give
them better result for certain applications are:
Fuzzy systems are suitable for uncertain or approximate
reasoning, particularly for a system with mathematical model,
which is difficult to derive.
Fuzzy logic allows decision building with estimated values under
incomplete or uncertain information.
In this section we have highlighted the problem which our customer
face while selecting the best possible combinations of product, the
PhD. Thesis by Devendera Agarwal 84
problem is because of the uncertainty in Semantic Web Taxonomies
[42]. Consider Indiatimes shopping portal shown in figure 5.1.
Fig. 5.1: Indiatimes Shopping Portal
If a buyer wants a laptop in the range of Rs.25000 < x < Rs.35000,
and with features F = {f1, f2, f3} in brands B = {b1, b2}, then he must
be shown the best possibilistic outcome of the above query.
The above problem looks very simple but it is not so, there exists an
uncertainty in the query, what if, if there is no laptop with all the
features of ‘F’ present in Brand ‘B’. Here comes a probabilistic
method to overcome such situation.
In our method, degrees of subsumption will be covered by Bayesian
Network based Ontology’s [30]. The Venn diagram is shown in figure
5.2.
PhD. Thesis by Devendera Agarwal 85
Fig. 5.2: Venn diagram Illustrating Electronic Items with Laptops
as one of their Categories & their Overlap
Our method enables the representation of overlap between a selected
concept and every other is referred taxonomy. The Price Range-I
represent the prices at the start of the price band while Price Range-II
represents the higher side of the price band.
The overlap is logic term expressed as
]1,0[|Referred|
|ReferredSelected|Overlap
(5.4)
The overlap region represents the value 0 for disjoint concepts and 1,
if the referred concept is subsumed by the selected one. This overlap
value can be used in information retrieval tasks. The match with the
query is generalized by the probabilistic sense and the hit list can be
sorted into the order of relevance accordingly.
If ‘F’ and ‘B’ are sets; then ‘F’ must be in one of the following
relationships to ‘B’.
PhD. Thesis by Devendera Agarwal 86
i) ‘F’ is a subset of ‘B’ i.e. . BF
ii) ‘F’ partially overlaps ‘B’ i.e.
)()(:, ByFyBxFxyx
iii) ‘F’ is disjoint from ‘B’ i.e. BF
Based on these relations we develop a simple transformation
algorithm. The algorithm processes the overlap graph G in a Breadth
First manner starting from root concept defined as ‘CON’. Each
processed concept ‘CON’ is written as the part of Solid Path
Structure (SPS).
if F subsumes B then
O := 1
else
C = Fs Bs
if C = then
O := 0
else
m(C)
O := m(B)Cc
end
end
Fig. 5.3: Computing the Overlap
PhD. Thesis by Devendera Agarwal 87
The overlap values ‘O’ for a elected concept ‘F’ and a referred concept
‘B’.
If F is the selected concept and B is referred one, then the overlap
value 0 can be interpreted as the conditional probability
0|(B)|s
s(B)S(F)true)F'|trueP(B'
(5.5)
where S(F) and S(B) are taken is and interpreted as a probability
space, and the elements of the sets are not interpreted as
elementary outcomes of some random phenomenon.
The implementation stages of the probabilistic search starts with the
Input of Ontology Rule which are refined in “Refinement Stage”. It is
than passed to the “Quantifier” which develops a set of Association
Rules. It is then fed to the further preprocessing by the “Naïve
Bayesian Transformation” module which finally generates the best
possible overlapping outcome as shown in figure 5.4.
Fig. 5.4: Implementation Framework.
PhD. Thesis by Devendera Agarwal 88
5.4 Conclusion
This chapter highlights the importance of Web Personalization and
Recommender Systems in e-commerce websites. In this chapter we
have highlighted how to identify a user having fraudulent intentions
using Naïve Bayes Classifier. Next to overcome the problems while
searching for a product on a portal we have proposed an algorithm
with implementation framework for selecting the best product fulfilling
user criterion. The model in both the cases uses Interactive Query
Refinement Mechanism to help to find the most suitable query terms.
The Ontology is planned according to restricted term relations. We
have developed an algorithm in which taxonomies can be constructed
without virtually any knowledge of Probability and Bayesian network.
The future extension could be to expand it using Fuzzy Regression
[128] with Bayesian Network.
***
PhD. Thesis by Devendera Agarwal 89
CHAPTER- 6
CONCLUSION
Chapter - 6
CONCLUSION
We would like to summarize our work by highlighting various features
of Web Personalization and Recommendation Model for Trust in
Ecommerce Website from an Indian Perspective.
We have studied that aptitude for growth of e-commerce in India is
colossal. Some findings like amount of awareness that is there for
online travel industry is not seen in case of other services. People
prefer non tangible goods i.e. services over tangible products.
Professional e-commerce websites are doing excellent job but still
there are some factors that are inhibiting users from purchasing
online. After survey and interview we concluded that for users to
adopt e-commerce, it is vital that the advantage of using this
commercial medium (e.g. convenience, saving in time and transaction
costs) considerably overshadow possible risks. Indeed, the consumer
freedom to choose suitable vendors needs to be correlated with greater
concerns regarding financial risk, privacy and trust. This can be
accounted for by the fact that private users are directly involved in the
commercial exchange; they are using their own equipment, giving
sensitive information about themselves as individuals, and spending
their own money. There are still various factors to be looked upon to
cater to needs of the consumer who is the driving force for e-
commerce. TRUST, CONVENIENCE, SECURITY are prime factors
without which we will not be able to attain our goal.
PhD. Thesis by Devendera Agarwal 90
In our quest for building Trust in “TrFRA: A Trust Based Fuzzy
Regression Analysis”, the major thrust of the model is to find out trust
building factors in e-commerce websites. It also focuses on fuzzy
relationship between Trust and website related factors. TrFRA is a
relationship model in terms of fuzzy output and regression line. The
output represents Trust Factor and its correlation with two factors
WOC (Web Object Content) and WOCA (Web Object Certification
Authority) also the estimated value for the output variable. TRUST is
derived by the difference between the observed and the estimated
value is assumed to be due to the ambiguity inherently present in the
system. The output Trust for a specified input is assumed to be a
range of possible values i.e. output can take on any of the possible
values. The advantage with Fuzzy Regression is the range of
Possibilistic values is much more as compared to the Normal
Regression Model.
Second phase of our work focuses on Trust on a website which leads
to increasing level of securities, this might irritant user enough to
migrate on other websites. In “Trust Vs Complexity of E-commerce
Sites”, we have highlighted how we need to tradeoff between Trust and
Complexity so as not to drive away users from the website. Our work
proposes a framework for measuring Trust on a website using
Classification Methodologies like Bayesian and Fuzzy Regression
Technique. The future extension of work is to find out the trade-off
between building Trust Vs Complexity. The major problem that we
PhD. Thesis by Devendera Agarwal 91
face when we try to make website trustworthy is its Complexity. As
Trust level increases security level also increases thereby increasing
the complexity/ interpretability.
To overcome the above situation we will be using Multi-objective
Optimization generating the Fuzzy Rule Base (FRBS) for Trust and
combine it with Evolutionary Multi-objective Optimization (EMO) to
generate the most optimal solution.
Secondly, we will be extending our work relating to Psychological
Aspects of Secured Transaction in Ubiquitous Environment. It will
cover the factor analysis of building TRUST in situations where mobile
devices are used for transaction.
Our work on “Web Personalization & Recommendation” highlights how
to identify fraud users by using Classification Methodologies of
Bayesian Rules and generating cluster of users having fraudulent
intentions. The model in both the cases uses Interactive Query
Refinement Mechanism to find the most suitable query terms. The
Ontology is planned according to restricted term relations. We have
developed an algorithm in which taxonomies can be created without
virtually any knowledge of Probability and Bayesian network. The
future extension could be to expand it using Fuzzy Regression with
Bayesian Network.
Most companies engaged in e-commerce employ various trust
promoting methods to convince consumers their websites are safe.
Even though these methods are universal, still many consumers are
afraid of carrying out transaction online. In this thesis a sincere effort
PhD. Thesis by Devendera Agarwal 92
has been made to understand why and how the degree to which trust
promoting methods can be employed to transform consumers’
attitudes toward online purchasing. Our findings will certainly help
shopping websites to work upon their existing trust promoting
methods and improve them by adding credible elements to enhance
the trust and their sales.
***
PhD. Thesis by Devendera Agarwal 93
REFERENCES
REFERENCES
1. A. Josang. (1998), “A Subjective Metric of Authentication”, 5th
European Symp. Research in Computer Security (ESORICS’98),
Springer-Verlag.
2. A. McCullagh (1998), “E-Commerce- A Matter of Trust”, the
Proceedings of the Information Industry Outlook Conference,
Canberra.
3. Aberg, J. and Shahmehri, N. (2000), “The Role of Human Web
Assistants in E-Commerce: An Analysis and a Usability Study.
Internet Research: Electronic Networking Applications and
Policy”, 10(2), 114-125.
4. ALEXA: The Web Information Company (2011), Website
www.alexa.com
5. All India Council of Technical Education (AICTE) (2011),
website www.aicte-india.org
6. Andrew S. Patrick (2002), “Building trustworthy Software
Agents”, IEEE Internet Computing, Vol 6, No. 6.
7. Audun Josang (1999), “An Algebra for Assessing Trust in
Certification Chains”, proceeding of the Network and
Distributed System Security (NDSS) Symposium.
8. Batya Friedman, Peter H. Kahn, JR., and Daniel. C. Howe
(2000), “Trust Online”, Communications of ACM, Vol.43, No.12,
pp34-40.
9. Bergenti, F. (2000), "Improving UML Designs using Automatic
Design Pattern Detection", in 12th International Conference on
Software Engineering and Knowledge Engineering (SEKE).
10. Bernstein, T., Bhimani A.B., Schultz, E. & Siegel, C. (1996),
“Internet Security for Business”, John Wiley, New York.
PhD. Thesis by Devendera Agarwal 94
11. Blaze, Feigenbaum and Keromytis (1998), “KeyNote: Trust
management for public-key infrastructures”, Lecture Notes in
Computer Science, 1550:59-63.
12. Blaze, Feigenbaum and Lacy (1996), “Decentralized trust
management.”, In Proceedings of the IEEE Symposium on
Research in Security and Privacy, Oakland, CA, pg 164.
13. Blaze, Feigenbaum, and Strauss (1998), “Compliance
checking in the PolicyMaker trust management systems”. In FC:
International Conference on Financial Cryptography. LNCS,
Springer-Verlag.
14. Camp, L. J. (2000), “Trust and Risk in Internet Commerce”,
The MIT Press.
15. Changa Hwang (2003), “Support Vector Fuzzy Regression
Machines”, In Transaction of Fuzzy Sets and Systems, Vol. 138,
Issue 2, ACM, USA.
16. Changsu, K., Jongheon, K., Namachul. S., Ryoo, J. H.
(2012), “Factors Influencing Internet Shopping Value and
Customer Repurchase Intention”, in Electronic Commerce
Research & Applications, Vol. 8, Issue 6.
17. Chen, L., Gillenson, M., & Sherrell, D. (2004), Consumer
Acceptance of Virtual Stores: A Theoretical Model and Critical
Success Factors for Virtual Stores. ACM SIGMIS Database, Vol.
35, Issue 2, pp 8-31.
18. Chen, Ming-Chung. (2007), “Mining User Progressive User
Behavior for E-Commerce using Virtual Reality Technique”, M.S.
Thesis, Faculty of Graduate School, University of Missouri-
Columbia.
19. Cheskin and Studio Archetype/Sapient (1999), “eCommerce
Trust Study”, www.cheskin.com/think/studies/ecomtrust.html.
PhD. Thesis by Devendera Agarwal 95
20. Ching-Chang Chen, Dong Her Shih, Hsiu-Sen Chiang, Che-
Hung Lin (2010), “An Empirical Study of Blog Marketing Based
on Trust and Purchase Intention”, International Review on
Computers and Software, Vol. 5, n. 1, pp. 97-105.
21. Cohen, R., Gorner, J., Zhang, J. (2012), “Improving Trust
Modeling through the Limit of Advisor Network size and use of
Referrals”, in Electronic Commerce Research & Applications,
Vol. 8, Issue 6.
22. Corder, G.W., Foreman, D.I. (2009), “Nonparametric Statistics
for Non-Statisticians: A Step-by-Step Approach”, Wiley.
23. Corritore, C.L., Widenbeck, S. and Kracher, B. (2001), “The
Elements of Online Trust”, Proceedings of CHI Conference on
Human Factors in Computing Systems, 504-505.
24. Csilla Farkas and Michael N. Huhns (2002), “Making Agents
Secure on the Semantic Web”, IEEE Internet Computing.
25. D. Dubois (1980), “Fuzzy Sets and Systems: Theory and
Applications”, Academic Press, USA.
26. D. Gambetta, Can We Trust Trust? (1988), chapter 13, pages
213{237. Basil Blackwell, Reprinted in electronic edition from
Department of Sociology, University of Oxford.
27. D.E. Goldberg (1989), “Genetic Algorithms in Search
Optimization and Machine Learning”, Addison Wesley, Reading,
MA.
28. D.Nauck, R.Kruse (1999), “Obtaining Interpretable Fuzzy
Classification Rules from Medical data”, Artificial Intelligence in
Medicine, Vol 16, Issue 2.
29. Daniel A. Menasce. (2002), “QoS Issues in Web Services”, IEEE
Internet Computing.
PhD. Thesis by Devendera Agarwal 96
30. Ding, Z. (2004), "A Probabilistic Extension to Ontology
Language OWL", in 12th Hawaii International Conference on
Systems Science.
31. Donna L. Hoffman, Thomas P. Novak, Marcos Peralta.
(1999), “Building Consumer Trust Online”, Communications of
ACM, Volume 42, Issue 4, pp 80-85.
32. E. L. Lehmann (1997), "Testing Statistical Hypotheses: The
Story of a Book". Statistical Science
33. Egger, F.N. (2000), “Trust Me, I'm an Online Vendor: Towards,
a Model of Trust for E-Commerce System design”, Proceeding of
the CHI2000 Conference on Human Factors in Computing
Systems, April, pp 101-102.
34. Eisenhauer, J.G. (2008), “Degrees of Freedom”, Teaching
Statistics, 30(3), 75–78
35. F. Azimzadeh, S.Khatun, B.M. Ali (2007), “A Method to
Incorporate Negative Value into a Trust Model”, International
Review on Computers and Software, Vol. 2, n. 2, pp. 190-197.
36. F. Hiller, Lieberman (1980), “Introduction to Operation
Research, IEEE Transaction on Fuzzy Systems”, IEEE Press,
New York.
37. Feldman Stuart (2000), “The Changing face of E-commerce:
Extending the Boundaries of the Possible”, IEEE Internet
Computing.
38. France Belanger, Varadharajan Sridhar, and Craig Van Slyke
(2002), “Comparing the Influence of Perceived Innovation
Characteristics and Trustworthiness Across Countries”,
Proceeding of the International Conference on Electronic
commerce Research (ICECR-5).
39. Glyn Carter (1998), “A Matter of Trust”, IEEE Review.
PhD. Thesis by Devendera Agarwal 97
40. Greenwood, P.E., Nikulin, M.S. (1996), “A guide to chi-
squared Testing”, Wiley, New York.
41. Grewal, H., Grewal S. (2012), “E-Commerce: Security
Challenges & Growth: An Indian Perspective”, in International
Journal of Management Sciences, Vol.1, Issue 2.
42. GuT (2004), “A Bayesian Approach for Dealing with Uncertain
Concepts”, in Conference Advances in Pervasive Computing,
Austria.
43. H. Ishibuchi, T. Yamamoto, T. Nakashima (2001), “Fuzzy
Data Mining: Effect of Fuzzy Discretization”, in: Proceedings of
the First IEEE International Conference on Data Mining, pp.
241–248.
44. H. Ishibuchi, Y. Nojima (2007), “Analysis of Interpretability–
Accuracy Tradeoff of Fuzzy Systems by Multiobjective Fuzzy
Genetics-Based Machine Learning”, International Journal of
Approximate Reasoning, Vol. 44, Issue 1, pp 4–31.
45. H. Wang, S. Kwong, Y. Jin, W. Wei, K.F. Man (2005), “Agent-
Based Evolutionary Approach for Interpretable Rule-Based
Knowledge Extraction”, IEEE Transactions on Systems Man,
and Cybernetics – Part C: Applications and Reviews, Vol. 35,
Issue 2, pp 143–155.
46. Haubl, G. & Trifts, V. (2000), Consumer Decision Making in
Online Shopping Environments: The Effects of Interactive
Decision Aids. Marketing Science, Vol. 19, Issue 1, pp. 4-21.
47. Huang, Y., Kuo, F. (2012), “How Impulsivity Affects Consumer
Decision Making in E-Commence”, in Electronic Commerce
Research & Applications, Vol.11 Issue 6.
48. Humpery Jonh, et.al. (2003), “The Reality of E-commerce with
developing Countries”, Report
PhD. Thesis by Devendera Agarwal 98
49. IAMAI & IMRB (2006/07), “Consumer E-commerce market In
India”, Report.
50. IAMAI & IMRB (2008), “I-CUBE 2008”, Report by IMRB
International, India.
51. IAMAI (2005), “Cybecafe Users: E-commerce Activites”, Report.
52. IAMAI (2007), “Annual Report 2006/07”, Report.
53. IAMAI (2007), “I-Cube 2007: Internet in India”, Report.
54. IAMAI (2008), “Annual Report 2007/08”, Report.
55. IAMAI (2010), “I-CUBE 2009-10”, Report by IMRB
International, India.
56. IOAI (2005), “IOAI Survey: Ecommerce Security 2005”, Survey
Report.
57. J. Lee, J. Kim, and J.Y. Moon (2000), “What Makes Internet
Users Visit Cyber Stores Again? Key Design Factors for
Customer Loyalty”, Proceedings of Human Factors in
Computing Systems, ACM press, pp. 305-312.
58. J. Yen, L. Wang (1998), “Application of Statistical Information
Criteria for Optimal Fuzzy Model Construction”, IEEE
Transactions on Fuzzy Systems, Vol. 6, Issue 3, pp 362–372.
59. J. Yen, L. Wang (1999), “Simplifying Fuzzy Rule-Based Models
Using Orthogonal Transformation Methods”, IEEE Transactions
on Systems, Man, and Cybernetics – Part B: Cybernetics, Vol.
29, Issue 1, 13–24.
60. J.C. Bezdek (1981), “Pattern Recognition with Fuzzy Objective
Function Algorithms”, Plenum Press, New York.
61. J.-S.R Jang, C. -T. Sun, E. Mizutani (1997), “Neuro-Fuzzy and
Soft Computing”, Prentice-Hall, Englewood Cliffs, NJ.
62. James. J. Buckley (2000), “Linear & Non Linear Fuzzy
Regression: Evolutionary Algorithms Solutions”, Fuzzy Sets and
PhD. Thesis by Devendera Agarwal 99
Systems, In Transaction of Fuzzy Sets and Systems, Vol. 109,
Issue 1, ACM, USA.
63. Jøsang A, and Pope S, (2005), ‘Semantic Constraints for Trust
Transitivity’ 2nd Asia-Pasific Conference on Conceptual
Modelling (APCCM2005), New Castle, Australia
64. Jøsang A, and Presti S, (2004), Analyzing the Relationship
between Risk and Trust, 2nd International Conference on Trust
Management.
65. Khaled S. Hassanein and Milena M. Head. (2002), “The
Impact of Humanized Web site Design on Establishing e-
Commerce Trust: A comparative Study across Different Product
Types”, Proceedings of International Conference on Electronic
Commerce Research (ICECR-5).
66. Koh Ai Tee. (2000), “E-Commerce in an Era of Creative
Destruction”, Available:
http://www.alumni.nus.edu.sg/almnus/jul2000/ecom.html.
67. KPMG (2008), “India Fraud Survey Report 2008”, KPMG
Forensic India.
68. L. Wang, J. Yen (1999), “Extracting Fuzzy Rules for System
Modeling Using a Hybrid of Genetic Algorithms and Kalman
filter”, Fuzzy Sets and Systems, Vol. 101, pp 353–362.
69. L.A. Zadeh (1972), “A Fuzzy-Set-Theoretic Interpretation of
Linguistic Hedges”, Journal of Cybernetics, Vol. 2, pp 4–34.
70. L.A. Zadeh (1973), “Outline of a New Approach to the Analysis
of Complex Systems and Decision Processes”, IEEE
Transactions on Systems, Man, and Cybernetics, Vol.3.
71. L.Castillo, A. Gonzalez, R.Perez (2001), “Including a
Simplicity Criterion in the selection of the Best Rule in a Genetic
Fuzzy Algorithm”, Fuzzy Sets and Systems, pp. 309-312.
PhD. Thesis by Devendera Agarwal 100
72. L.F. Cranor, J. Reagle and M.S. Ackerman (1999), “Beyond
Concern: Understanding Net Users’ Attitudes About Online
Privacy”, AT&T Labs-Research Technical Report TR 99.4.3.
73. Lee. Ming-Ling (2003), “Modeling of Hierarchical Fuzzy
Systems, In Transaction of Fuzzy Sets and Systems”, Vol. 138,
Issue 2, ACM, USA.
74. Lehmann, E.L.; Joseph P. Romano (2005), “Testing Statistical
Hypotheses” (3E ed.). New York: Springer.
75. Lenard Huff, Lane Kelley (2003), “Levels of Organizational
Trust in Individualist versus Collectivist Societies: A Seven-
Nation Study”. Organization Science, Vol. 14, No. 1, pp. 81-90.
76. M. Daignault, M. Shepherd, S. Marche and C. Watters
(2001), “Enabling Trust Online”,
http://ecommerce.ncsu.edu/ISEC/papers/01_shepherd_enabli
ng.pdf.
77. M. Darbari, B. Karn, V. Kr. Singh, S. S. Ahmad (2008),
“Integrating Natural Language Requirements and OPEN Process
Activity Theory Model for Platform Independent Web
Monitoring”, International Review on Computers and Software,
Vol. 3, n3, pp. 283-287.
78. M. J. Kargar, A. R. Ramli, H. Ibrahim, S.B. Noor (2007),
“Towards a Practical and Valid Model for Assessing Quality of
Information on the Web”, International Review on Computers
and Software, Vol. 2, n. 2, pp. 80-88.
79. M. Sabeghi, M. Naghibzadeh (2006), “A Fuzzy Algorithm for
Real-Time Scheduling of Soft Periodic Tasks”, International
Review on Computers and Software, Vol. 1, n. 2, pp. 106-113.
80. M. Setnes, H. Roubos (2000), “GA-Fuzzy Modeling and
Classification: Complexity and Performance”, IEEE Transactions
on Fuzzy Systems, Vol. 8, Issue 5, pp 509–522.
PhD. Thesis by Devendera Agarwal 101
81. M. Setnes, R. Babuska, B. Verbruggen (1998), “Rule-Based
Modeling: Precision and Transparency”, IEEE Transactions on
Systems, Man, and Cybernetics – Part C 28, Vol. 1, pp 165–169.
82. M. Sugeno, T. Yasukawa (1993), “A Fuzzy-Logic-Based
Approach to Qualitative Modeling”, IEEE Transactions on Fuzzy
Systems, Vol. 1, pp 17–31.
83. M.J. Gacto, R. Alcala, F. Herrera (2009), “Adaptation and
Application of Multi-objective Evolutionary Algorithms for Rule
Reduction and Parameter Tuning of Fuzzy Rule Based Systems”,
Soft Computing, Vol. 13, Issue 5, pp 419–436.
84. Massimo Marchiori (1998), “The limits of Web metadata and
beyond”, The World Wide Web Consortium (W3C), MIT
Laboratory for Computer Science, USA.
85. McKnight D.H., and Chervaney N.L. (1996), ‘The Meanings of
Trust, Technical Report MISRC Working Paper Series 96-04,
University of Minnasota, Management Information Systems
Research Center
86. Milena M. Head, Khaled Hassanein, and Edword Cho. (2002),
“Online Trust through Humanized Web site Design: Research
Model and Empirical Study”, Proceedings of the 5th
International Conference on Electronic Commerce Research,
Montreal, Quebec.
87. N. M. Frank and L. Peter (1998), “Building Trust: the
Importance of Both Task and Social Precursors”, International
conference of Engineering and Technology Management:
Pioneering New Technologies – Management Issues and
Challenges in the Third Millennium, URL at
http://ieeexplore.ieee.org/iel4/5884/15675/00727781.pdf.
88. O. Cordon, F. Herrera, F. Hoffmann, L. Magdalena (2001),
“Genetic Fuzzy Systems –Evolutionary Tuning and Learning of
Fuzzy Knowledge Bases”, World Scientific, Singapore.
PhD. Thesis by Devendera Agarwal 102
89. P. Lamsal, “Understanding Trust and Security (2001)”,
http://www.cs.Helsinki.FI/u/lamsal/papers/UnderstandingTru
stAndSecurity.pdf.
90. Paul Resnick and James Miller (1996), “PICS: Internet Access
controls Without Censorship”, Communications of the ACM,
vol39 (10), pp.87-93.
91. Paul Resnick, Richard Zeckhauser, Eric Friedman, and Ko
Kuwabara (2000), “Reputation Systems”, Communications of
the ACM, Vol.43, No.12, pp45-48.
92. Peter Keen. (1999), “Electronic Commerce and the Concept of
Trust”, URL at www.peterkeen.com/ecr2.htm.
93. Peter Kollock (1999), “The Production of Trust in Online
Markets”, Advances in Group Processes/vol.16, edited by E. J.
Lawler, M. Macy, S. Thyne, and H. A. Walker. Greenwich,
CT:JAI Press.
94. PEW/Internet (2008), “Online Shopping”, Survey Report.
95. PM, Doney, JP, Cannon, M. Mullen (1998), “Understanding
National Culture on the development of Trust”, Academy of
Management Review, pp 601-620.
96. Princeton Survey Research Associates. (2002), “A Matter of
Trust: What Users Want from Web sites”,
http://www.consumerwebwatch.org/news/report1.pdf.
97. Pwc Annual Report (2011), “Emerging Opportunities –
Financial Services M&A in Asia 2011”, Available at
http://www.pwc.com/gx/en/mergers-acquisitions-industry-
trends/survey/index.jhtml
98. R. Alcala, P. Ducange, F. Herrera, B. Lazzerini, F. Marcelloni
(2009), “A Multiobjective Evolutionary Approach to
Concurrently Learn Rule and Databases of Linguistic Fuzzy-
PhD. Thesis by Devendera Agarwal 103
Rule-Based Systems”, IEEE Transactions on Fuzzy Systems,
Vol. 17, Issue 5, pp 1156–1122.
99. R. Shrestha (2010), “Fuzzy Nonlinear Regression Approach to
Stage-Discharge Analysis: Case Study”, Journal of Hydrology
Engineering, USA.
100. Rachrela, P., Friske, W. (2012), “Perceived ‘usefulness’ of
Online Consumer Reviews: An Exploratory Investigation across
Three Services Categories”, in Electronic Commerce Research &
Applications, Vol.11 Issue 6.
101. Rastogi, Rajiv (2007), “INDIA: Country Report on E-Commerce
Initiatives”.
102. Ravi Kalakota , Andrew B. Whinston (1996), “Frontiers of
electronic commerce”, Addison Wesley Longman Publishing Co.,
Inc., Redwood City, CA
103. Ravi Kalakota, Andrew B. Whinston (1997), “Electronic
Commerce: A Manager’s Guide”, Addison Wesley.
104. Ravi Sandhu (2002), “The Technology of Trust”, The IEEE
Internet Computing.
105. Robert R. Hoffman. (1998), “Whom (or What) Do You
(Mis)Trust?: Historical Reflections on the Psychology and
Sociology of Information Technology”, Fourth Annual
Symposium on Human Interaction with Complex Systems.
IEEE.
106. Rohit Khare (1997), “Digital Signature Label Architecture”,
URL at http://www.w3.org/pub/WWW/TR/WD-DSIG-label-
arch-970610.html.
107. Rojas, H. Pomares, J. Ortega, A. Prieto (2000), “Self-
organized Fuzzy System Generation from Training Examples”,
IEEE Transactions on Fuzzy Systems, Vol. 8, Issue 1, pp 23–36.
PhD. Thesis by Devendera Agarwal 104
108. Schafer, J. B., Konstan, J. A., & Riedl, J. (1999),
“Recommender Systems in E-Commerce”. In EC ’99:
Proceedings of the First ACM Conference on Electronic
Commerce, Denver, CO., pp 158-166.
109. Schafer, J. B., Konstan, J. A., & Riedl, J. (2001), “E-
Commerce Recommendation Applications. Data Mining and
Knowledge Discovery”, Vol. 5, Issue 1/2, pp115-153.
110. Schafer, J.Ben. (2008), et. al., “E-Commerce Recommendation
Applications”, as Grouplens Research Project, University of
Minnesota.
111. Senecal, S. & Nantel, J. (2004), “The Influence of Online
Product Recommendations on Consumers”, Online Choices.
Journal of Retailing, Vol. 80, pp 159–169.
112. Smith, D., Menon, S., & Sivakumar, K. (2005), “Online Peer
and Editorial Recommendations, Trust, and Choice in Virtual
Markets”. Journal of Interactive Marketing, Vol. 19, pp 15-37.
113. T. Grandison and M. Sloman (2000), “A survey of Trust in
Internet Applications”, IEEE Communications Surveys, 4th
quarter.
114. T. Takagi, M. Sugeno (1985), “Fuzzy Identification of Systems
and its Applications to Modeling and Control”, IEEE Transaction
Systems, Man and Cybernetics, Vol.15, Issue 1, pp 116–132.
115. T.A. Johansen, R. Babuska (2003), “Multiobjective
Identification of Takagi-Sugeno Fuzzy Models”, IEEE
Transactions on Fuzzy Systems, Vol. 11, Issue 6, pp 847–860.
116. Thomas P. Novak, Donna L. Hoffman and Marcos Peralta
(1999), “Building Consumer Trust in Online”, Communications
of the ACM, vol42, No.4, pp80-85.
PhD. Thesis by Devendera Agarwal 105
117. Trevor Hastie, Robert Tibshirani, Jerome H. Friedman
(2009), “The elements of statistical learning: data mining,
inference, and prediction”, Springer 2nd ed., 746 p.
118. W. Nather (2003), “Least Squares Fuzzy Regression with Fuzzy
Random Variables”, In Transaction of Fuzzy Sets and Systems,
Vol. 130, Issue 1, ACM, USA.
119. W3C (1998), “REFEREE Project”, URL at
http://www.w3.org/PICS/TrustMgt/.
120. Waghmare, G.T. (2012), “E-Commerce; A Business Review &
Future Prospects in Indian Business”, in Indian Streams
Research Journal, Vol.2, Issue IV.
121. Wang, Lipo. (2006), “Fuzzy Systems & Knowledge Discovery”,
Springer.
122. Wigand, R. T (1997), “Electronic Commerce: Definition, Theory
and Context, The Information Society”, Vol 13, 1-16
123. Wikipedia : The Free Encyclopedia (2011), “Introduction
about India” Available: http://en.wikipedia.org/wiki/India
124. Y. Jin (2000), “Fuzzy Modeling of High-Dimensional Systems:
Complexity Reduction and Interpretability Improvement”, IEEE
Transactions on Fuzzy Systems, Vol. 8, Issue 2, pp 212–221.
125. Y. Jin, W. von Seelen, B. Sendhoff (1999), “On Generating
FC3 Fuzzy Rule Systems from Data Using Evolution Strategies”,
IEEE Transactions on Systems, Man, and Cybernetics – Part B:
Cybernetics, Vol. 29, Issue 6, pp 829–845.
126. Y. Yoshinari, W. Pedrycz, K. Hirota (1993), “Construction of
Fuzzy Models Through Clustering Techniques”, Fuzzy Sets and
Systems, Vol. 54, pp 157–165.
127. Yinan Yang, Lawrie Brown, Ed Lewis (2001), “eCommerce
Trust via the Proposed W3 Model”, Published in proceedings of
PhD. Thesis by Devendera Agarwal 106
postgraduate ADFA conference on computer science, University
of New South Wales, Australia.
128. Zongmin, Ma. (2006), “Soft Computing in Ontologies and
Semantic Web”, Springer.
129. Zwass. V (1998), “Structure and Macro-Level Impacts of
Electronic Commerce: From Technological Infrastructure to
Electronic Market Places in Emerging Information Technologies”
ed. Kenneth E. Kendall, Thousand Oaks, CA: Saga Publications.
PhD. Thesis by Devendera Agarwal 107
NOTES
PhD. Thesis by Devendera Agarwal 108
APPENDIX
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 6, November 2010 ISSN (Online): 1694-0814 www.IJCSI.org
18
Web Personalization of Indian e-Commerce Websites using Classification Methodologies
Agarwal Devendera1, Tripathi S.P2 and Singh J.B.3
1 School of Computer Science & Information Technology, Shobhit University, Research Scholar Meerut, U.P. 250110, India
2 Department of Computer Science & Engineering, Gautam Buddh Technical University, IET Lucknow, U.P. 226021, India
3 School of Computer Science & Information Technology, Shobhit University Meerut, U.P. 250110, India
AbstractThe paper highlights the classification methodologies using Bayesian Rule for Indian e-commerce websites. It deals with generating cluster of users having fraudulent intentions. Secondly, it also focuses on Bayesian Ontology Requirement for efficient Possibilistic Outcomes. Keywords: E-commerce, Possibilistic Outcome, Bayesian Rule, Ontology.
1. Introduction
Electronic Commerce is fast emerging as most popular method of purchasing, let it be a small pen drive or bulky LED TV. Recent survey [3] has estimated that around 3-5% of Indians have transacted or are well versed with working of online shopping websites. The strategy which is being followed until now related to the various policy initiatives like:
Consumer Proportion: This model is being propagated by the government based on certain guidelines for the protection of consumers. Legality: It deals with formal recognition of electronic signatures; In India digital signatures are necessary for e-Tendering. Security: Central Government has issued its policy relating to cryptography techniques to ensure secure electronic commerce in third party transfer.
In order to deal with security and web personalization [2] issues we develop two basic classification methods: Naïve Bayes and K-nearest neighbor.
2. Our Model
In order to make our model more illustrative we are taking example of “Predicting Fraudulent Transaction”.An Indian e-commerce company has a very large customer base; each customer has to submit his personal information before making a transaction. In this way each company is acting as a record and the response of internet is given as
Z = {Fraudulent, Trustworthy} (1) these are the classification in which we can categorize a customer. By analyzing from a sample e-commerce site we are able to find out that in case of Fraudulent the customer-id should be reposted to the e-fraud cell. Two set of data are taken to check the consistency of data.
Table 1: Report of Customer on e-commerce site.
Reportingto e-fraud
cell
No Reporting Required Total
fraudulent 20 80 100 trustworth
y 100 300 400 Total 120 380 500
2.1 Naïve Bayes
In order to classify record into ‘m’ classes by ignoring all predictor information X1, X2,….., Xp is to classify the record as a member of majority class. For example in our case naïve rule would classify all the customers to be
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 6, November 2010 ISSN (Online): 1694-0814 www.IJCSI.org
19
“Trustworthy”, because 90% of the companies were found to be Truthful.Naïve Bayes classifier [1] is an advanced version of Naïve rule. The logic to introduce Bayes is to integrate the information given in the set of predictors into the naïve rule to obtain more accurate classifications. The methodology suggests in finding out the probability of record belonging to a certain class is evaluated on the prevalence of that class along with additional information that is being given on that record in terms of Xinformation. Since our dataset is very large we prefer Naïve Bayes method. In a classification task our goal is to estimate the probability of membership to each class given a certain set of predictor variables. This type of probability is called a conditional probability. In our example we are interested in P (Fraudulent | Reporting to e-fraud cell). In general, for a response of ‘m’ classes C1, C2, ….., Cm and the predictors X1, X2, ….., Xp we compute as:
P (Ci | X1,…,Xp) where i = 1, 2 , …, m. (2)When the predictors are all categorical we can use a pivot to estimate the confidential probabilities of class membership. Consider its application in our example we compute the probabilities divided into two classes as: For P (Fraudulent | Reporting to e-fraudulent cell) = 20/120andP (Trustworthy | Reporting to e-fraudulent charges) = 100/120.The above statement indicates that although the firm is still more likely to be Trustworthy than Not Trustworthy,the probability of its being Truthful is much lower than the naïve rule. However, the method usually gives good result partly because what is important is not the exact probability estimate but the ranking for that case in comparison to others.In order to convert the desired probabilities into class probability we use Bayes Theorem. The Bayes Theorem gives us the following formula to compute the probability that the record belongs to class Ci:
)()|,...,(...)()|,...,()()|,...,(
),...,|(1111
1111
mmPP
PPi CPCXXPCPCXXP
CPCXXPXXCP (3)
Ci: To compute the numerator we filter two pieces of information i) The proportion of each class in the population
[P(C1)……P(Cm)]ii) The probability of occurrence of the predictor
vales X1, X2, …, Xp within each class from the training set.
We develop another table of the User which is categorized as “Frequent Buyers” and “Occasional Buyers”, for each of these two categories of Buyers we have information on whether or not reporting has been done,
and whether it turned out to be Fraudulent or Trustworthy.
Table 2: Sample of 10 users.
Reporting to e-fraud cell User-Type Status
Yes Occasional Buyer Fraudulent No Occasional Buyer Trustworthy No Frequent Buyer FraudulentNo Frequent Buyer TrustworthyNo Occasional Buyer TrustworthyNo Occasional Buyer TrustworthyNo Frequent Buyer TrustworthyYes Occasional Buyer FraudulentYes Frequent Buyer FraudulentNo Frequent Buyer Fraudulent
The probability of fraud can be defined by four possible states {Yes, Occasional Buyer}, {Yes, Frequent Buyer},{No, Occasional Buyer}, {No, Frequent Buyer}.
i) P(Fraudulent | Reporting = Yes, Customer Type = Occasional Buyer) = 1/2 = 0.5
ii) P(Fraudulent | Reporting = Yes, Customer Type = Frequent Buyer) = 2/2 = 1
iii) P(Fraudulent | Reporting = No, Customer Type = Occasional Buyer) = 0/3 = 0
iv) P(Fraudulent | Reporting = No, Customer Type = Frequent Buyer) = 1/3 = 0.33
We can extend this for Naïve Bayes probabilities, for analyzing the conditional probabilities of fraudulent behavior “Reporting to e-fraud cell” = Yes, and “UserType” = Occasional Buyer, the numerator is a proportion of “Reporting to e-fraud cell”. Instances amongst the type of Buyers, times the proportion of Fraudulent Customers
= (3/4) (1/4) (4/10) = 0.075 To get the actual probability we calculate the numerator for the conditional probability of truth given
Reporting to e-Fraudulent Cell = Yes; Type of Customer = Occasional Buyer;
The denominator is then the sum of two conditional probabilities
= (0.075 + 0.067) = 0.14 Therefore the conditional probability of fraudulent behaviors is given by
PNB(Fraudulent | Reporting to e-Fraudulent cell = Yes; Buyer Type = Occasional )
= (3/4)(1/4)(4/10) (3/4)(1/4)(4/10)+(1/6)(4/6)(6/10)
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 6, November 2010 ISSN (Online): 1694-0814 www.IJCSI.org
20
= 0.075/0.14 = 0.53 PNB(Fraudulent | Reporting to e-Fraudulent cell = Yes; Buyer Type = Frequent ) = 0.087 PNB(Fraudulent | Reporting to e-Fraudulent cell = Yes; Buyer Type = Occasional ) = 0.031
Rank Ordering of probabilities are even closer to exact Bayes method than are the probabilities themselves, to further analyze we can use classification matrix.
2.1 Advantages & Disadvantages of Naïve Bayes Classifier
The logic of using Naïve Bayes Classification Technique [7] is to attain computational efficiency and good performance.
2.2 Fuzzy Information Classification and Retrieval Model
The above section deals with a classification technique [6] by which we can categorize the customer visiting our site based on their transaction history. In this section we have highlighted the problem which our customer face while selecting the best possible combinations of product, the problem is because of the uncertainty in Semantic Web Taxonomies [8]. Consider Indiatimes shopping portal shown in fig. 1.
Fig. 1: Indiatimes Shopping Portal.
If a buyer wants a laptop in the range of Rs.25000 < x < Rs.35000, and with features F = {f1, f2, f3} in brands B = {b1, b2}, then he must be shown the best possibilistic outcome of the above query. The above problem looks very simple but it is not so, there exists an uncertainty in the query, what if, if there is no laptop with all the features of ‘F’ present in Brand ‘B’. Here comes a probabilistic method to overcome such situation.
In our method, degrees of subsumption will be covered by Bayesian Network based Ontology’s [4]. The Venn diagram shown in figure 2
Fig. 2: Venn Diagram Illustrating Electronic Items with Laptops as one of their Categories & their Overlap.
Our method enables the representation of overlap between a selected concept and every other is referred taxonomy. The Price Range-I represent the prices at the start of the price band while Price Range-II represent the higher side of the price band. The overlap is logic term expressed as
]1,0[|Referred|
|ReferredSelected|Overlap (4)
The overlap region represents the value 0 for disjoint concepts and 1, if the referred concept is subsumed by the selected one. This overlap value can be used in information retrieval tasks. The match with the query is generalized by the probabilistic sense and the hit list can be sorted into the order of relevance accordingly. If ‘F’ and ‘B’ are sets; then ‘F’ must be in one of the following relationships to ‘B’.
i) ‘F’ is a subset of ‘B’ ie BF .ii) ‘F’ partially overlaps ‘B’ ie
)()(:, ByFyBxFxyxiii) ‘F’ is disjoint from ‘B’ ie BF
Based on these relations we develop a simple transformation algorithm. The algorithm processes the overlap graph G in a Breadth First manner starting from root concept defined as ‘CON’. Each processed concept ‘CON’ is written as the part of Solid Path Structure (SPS).The overlap values ‘O’ for a elected concept ‘F’ and a referred concept ‘B’
ElectronicItems
Laptopsf1
f2
f3
B1
B2
B3
Price Range - I
Price Range - II
Mutual Overlap atB2 Brand
IJCSI International Journal of Computer Science Issues, Vol. 7, Issue 6, November 2010 ISSN (Online): 1694-0814 www.IJCSI.org
21
Fig. 3: Computing the Overlap.
If F is the selected concept and B is referred one, then the overlap value 0 can be interpreted as the conditional probability
0|(B)|s
s(B)S(F)true)F'|trueP(B'
where S(F) and S(B) are taken is and interpreted as a probability space, and the elements of the sets are not interpreted as elementary outcomes of some random phenomenon. The implementation stages of the probabilistic search starts with the Input of Ontology Rule which are refined in “Refinement Stage”. It is than passed to the “Quantifier”which develops a set of Association Rules. It is then fed to the further preprocessing by the “Naïve Bayesian Transformation” module which finally generates the best possible overlapping outcome as shown in figure 4.
3. Conclusions & Future Scope
The model in both the cases uses interactive query refinement mechanism to help to find the most appropriate query terms. The Ontology is organized according to narrower term relations. We have developed an algorithm in which taxonomies can be constructed without virtually any knowledge of Probability and Bayesian network. The future extension could be to expand it using Fuzzy Regression [7] with Bayesian Network.
Acknowledgments
Agarwal, Devendera thanks Darbari, Manuj without his essential guidance this research paper would not have been possible and also to management of Goel Group of
Institutions for their financial support in his maiden endeavor.
Fig. 4: Implementation Framework.
References[1] Bergenti, F., "Improving UML Designs using Automatic
Design Pattern Detection", in 12th International Conference on Software Engineering and Knowledge Engineering (SEKE), 2000.
[2] Chen, Ming-Chung., “Mining User Progressive User Behavior for E-Commerce using Virtual Reality Technique”, M.S. Thesis, Faculty of Graduate School, University of Missouri-Columbia, 2007.
[3] IAMAI, “I-CUBE 2009-10”, Report by IMRB International, India, 2010.
[4] Ding, Z., "A Probabilistic Extension to Ontology Language OWL", in 12th Hawaii International Conference on Systems Science, 2004.
[5] GuT, “A Bayesian Approach for Dealing with Uncertain Concepts”, in Conference Advances in Pervasive Computing, Austria, 2004.
[6] Schafer, J.Ben., et. al., “E-Commerce Recommendation Applications”, as Grouplens Research Project, University of Minnesota, 2008.
[7] Wang, Lipo., “Fuzzy Systems & Knowledge Discovery”, Springer, 2006.
[8] Zongmin, Ma., “Soft Computing in Ontologies and and Semantic Web”, Springer, 2006.
Agarwal, Devendera is currently working as Prof. & Director (Academics) at Goel Institute of Technology & Management, Lucknow. He has over 12 years of teaching & 5 years of industrial experience. Having done his B.Tech in Computer Science from Mangalore University in 1993, M.Tech from U.P.Technical University, Lucknow in 2006, he is pursuing his Ph.D. from Shobhit University, Meerut.
Tripathi, S.P. (Dr.) is currently working as Assistant Professor in Department of Computer Science & Engineering at I.E.T. Lucknow. He has over 28 years of experience. He has published numbers of papers in referred National Journals.
Singh, J.B. (Dr.) is currently working as Dean Students Welfare at Shobhit University, Meerut. He has 38 years of teaching experience and has published number of papers in referred National Journals.
Raw Ontology File Structure
RefinementStage Quantifier
Naïve Bayesian Transformation
Best Possible Outcome
if F subsumes B then O := 1 else C = FS BS
if C = then O : = 0
else m(C)
O: = m(B)
Cc
endend
International Journal of Scientific & Engineering Research Volume 3, Issue 4, April-2012 1ISSN 2229-5518
IJSER © 2012 http://www.ijser.org
Trust Vs Complexity of E-Commerce Sites Devendera Agarwal, R.P.Agarwal, J.B.Singh, S.P.Tripathi
Abstract— E-Commerce suffers from uncertainty which can produce devastating results. The user first checks the level of security and then proceeds further. At the same time the user switches to another e-commerce site if he has to deal with several layers of security. To overcome this drawback e-commerce sites are now finding a solution of maintaining high security (Trust) with lesser complexity as far as possible. Our paper focuses on the issue of development of a framework to provide an optimal relationship between the two.
Index Terms— Complexity, Threat to e-commerce, Fuzzy Rule, Security, Tradeoff, Transaction, Trust.
—————————— ——————————
1 INTRODUCTION
NDIA today is facing with various kinds of threat to e-commerce systems. The problem arises when we increase the security of the e-commerce website, the complexity at
the user level also increases, which in turn affects the volume of sale. While traditional marketing does not involve any type of complexity since the consumer deals directly with the sup-plier. Since internet marketing does not involve any face to face direct interaction so a visual interface is essential. There are various types of online buying behavior models like Bett-man (1979) and Booms (1981) in which the focus was on per-sonal characteristics viz. Culture, Social Group and Physiolog-ical Behavior. Lewis and Lewis (1997) identified five different types of web which remain valid today:
Directed information-seekers: These users will be look-ing for product information and are not typically planning to buy online. Undirected information-seekers: These are the users, usually referred to as 'surfers', who like to browse and change sites by following hyperlinks. Members of this group tend to be novice users and may also click banners of the website. Directed buyers: These buyers are online to purchase spe-cific products online. For such users, brokers that compare product features and prices will be important locations to visit. Bargain hunters: These users want to find the offers available from sales promotions such as free samples or competitions. Entertainment seekers: These are users looking to interact with the Web for enjoyment through entering contests such as quizzes, puzzles or interactive multi-player games.
Under all the above categories the main focus is the trust of web users [24], [25] which will finally lead to purchase. The communication between server and client are not secure un-
less it is providing a safe and secure transaction. To reduce the risk we have to deal with development of trustworthiness of the web services, which finally means increasing the complexi-ty of the website.
From the above figure we can conclude that there has to be some situation in which a trade off between Trust Level and Complexity of the transaction has to be maintained. This trade off can be achieved by the help of development of Fuzzy Rule base, but simple Fuzzy Rule base will not be sufficient for this purpose, so we extend this problem and solve it using Evolu-tionary Multi-objective Optimism [9], [10], [12].
2 LITERATURE SURVEY
The work by H. Ishibuchi & H. Tanaka (1994) highlights the construction of Fuzzy Classification of various entities using genetic Algorithms. Later on they extended their work (1995) using If-Then-Else rules. M. Setnes (1998) developed a Rule-Based system for developing the Precision & Transparency. D. Nauck (1999) worked on the interpretability aspect of Medical Data; we are motivated by their work and extending it for e-commerce websites. Y. Jin (2000) developed a framework for modeling high dimensional system in finding out their Com-plexity and Interpretability aspect. L.Castillo (2001) developed the best rule in a genetic fuzzy learning algorithm. M.Setnes (2000) also developed a mechanism of GA-based Modeling & Classification which measures the Complexity and Perfor-
I
————————————————
Devendera Agarwal is currently pursuing his Ph.D. from Shobhit Univer-sity, Meerut, India. E-mail: [email protected]. R.P.Agarwal & Prof. J.B.Singh, Shobhit University, Meerut, India, E-mail: [email protected], [email protected]. S.P. Tripathi, Department of Computer Science, I.E.T, Lucnkow, India, E-mail: [email protected]
LOW HIGH
HIGH
LOW
If TRUST level is HIGH and COMPLEXITY is LOW consu mer will PREFER this.
If TRUST level is HIGH and COMPLEXITY is H IGH consumer will RESIGN .
If TRUST level is LOW then COMPLEXITY has to be LOW consumer will AVOID this.
If TRUST level is LOW and COMPLEXITY is H IGH consumer will ABANDON .
Fig. 1. Trust/ Complexity Matrix.
International Journal of Scientific & Engineering Research Volume 3, Issue 4, April-2012 2ISSN 2229-5518
IJSER © 2012 http://www.ijser.org
mance of the system.
3 OUR PROPOSED MODEL
Genetic Algorithms [6], [8], [17] have been frequently used to model a solution for conflicting goals. Let Trust (T) be a measure of security which the customer will be provided and Inverse of Complexity (C) be the user comfort level. Applying the Fuzzy Rule base we can get
Maximize Trust (T) (1)
But it leads to compromise in the complexity (C) of fuzzy rule based systems [2], [3], [5], [7]. According to consumers survey most of consumers in India considers Trust and Ease of Use (Lower level of Complexity) at the same time. The above problem can be formulated as
Maximize Trust (T) subject to Inverse complexity (C) (2)
where complexity (C) is the measure of fuzzy rule system.
We can develop a single objective function to the above so-lution given as:
Maximize ƒ(Trust (T), Inverse of Complexity (C)) (3)
We can also use weights in order to determine the exact function for e-commerce site.
Maximize (w1) Trust (T) + (w2) Inverse of Complexity (C) (4)
We proceed with development of more refined stages in which we can focus on various stages of membership func-tions. Consider a simple single output function y = f(x) an ap-plication of Takagi-Sugeno method [7], [11], [15] we can write it as:
Rule Ri : if x is Ai then y=ai+bjx, i=1,2,…N Rule Rk : if x is Ak then y=ak+bkx, k=1,2,…N ::Rule Rz : if x is Az then y=az+bzx, z=1,2,…N (5)
This output value is given as:
N
iAi
Ai
N
ixi
x
xbaxy
1
1
)(
)()()( (6)
where y(x) is the estimated output value for the input value x i (x) is the membership value of the antecedent fuzzy
set Ai.
From the input-output data we can derive the relationship between Trust and Complexity of the e-commerce site consi-dering three Sugeno Rules.
We develop a heuristic rule[1], [2] denoted by three lines A, B and C as the subsequent of the linear function with fuzzy sets A1, A2 and A3. Each of the Fuzzy Rule can be represented in triangular Fuzzy Sets.
Rule R1: If TRUST is SMALL and COMPLEXITY is HIGH Then User’s Ease of Use is MEDIUM.
Rule R2: If TRUST is LARGE and COMPLEXITY is ME-DIUM Then Users Ease of Use is HIGH
Rule R3: If TRUST is SMALL and COMPLEXITY is SMALL Then Users Ease of Use is HIGH
Based on the above rules we try to develop a plot between Complexity and Trust and develop our interpretable solution [13], [14], 16], [18], [19] between the two entities.
Possibly we can also merge the above set of rules to achieve more refined results, but a relationship generated by optimiza-tion rules gives some gridlines in the area of relationship be-tween the two entities.
A B C
Fig. 2. Three Takagi-Sukeno Rules
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Trust
Com
plex
ity
Fig. 3. Input Output Data using Fuzzy Data Set
International Journal of Scientific & Engineering Research Volume 3, Issue 4, April-2012 3ISSN 2229-5518
IJSER © 2012 http://www.ijser.org
4 CONCLUSIONIt is very difficult to interpret the exact relationship be-
tween the two entities. Different Fuzzy rule are being applied in order to determine the appropriate interpretability. The method that we have used is the application of Fuzzy Optimi-zation Theory [20], [21], [22], [23] to find the probable relation-ship between Complexity and Trust. The future extension would be to use Evolutionary Algorithm [4], [5] in finding out the best possible trade-off between the two entities.
REFERENCES
[1] I. Rojas, H. Pomares, J. Ortega, A. Prieto, “Self-organized Fuzzy System Gen-eration from Training Examples”, IEEE Transactions on Fuzzy Systems 8 (1) (2000) 23–36.
[2] T. Takagi, M. Sugeno, “Fuzzy Identification of Systems and its Applications to Modeling and Control”, IEEE Transaction Systems, Man and Cybernetics 15 (1) (1985) 116–132.
[3] J.C. Bezdek, “Pattern Recognition with Fuzzy Objective Function Algo-rithms”, Plenum Press, New York, 1981.
[4] L.A. Zadeh, “Outline of a New Approach to the Analysis of Complex Sys-tems and Decision Processes”, IEEE Transactions on Systems, Man, and Cy-bernetics 3 (1973) 28–44.
[5] J.-S.R Jang, C. -T. Sun, E. Mizutani, “Neuro-Fuzzy and Soft Computing”, Prentice-Hall, Englewood Cliffs, NJ, 1997.
[6] O. Cordon, F. Herrera, F. Hoffmann, L. Magdalena, “Genetic Fuzzy Systems –Evolutionary Tuning and Learning of Fuzzy Knowledge Bases”, World Scientific, Singapore, 2001.
[7] Y. Jin, W. von Seelen, B. Sendhoff, “On Generating FC3 Fuzzy Rule Systems from Data Using Evolution Strategies”, IEEE Transactions on Systems, Man, and Cybernetics – Part B: Cybernetics 29 (6) (1999) 829–845.
[8] H. Ishibuchi, Y. Nojima, “Analysis of Interpretability–Accuracy Tradeoff of Fuzzy Systems by Multiobjective Fuzzy Genetics-Based Machine Learning”, International Journal of Approximate Reasoning 44 (1) (2007) 4–31.
[9] M.J. Gacto, R. Alcala, F. Herrera, “Adaptation and Application of Multi-objective Evolutionary Algorithms for Rule Reduction and Parameter Tuning of Fuzzy Rule Based Systems”, Soft Computing 13 (5) (2009) 419–436.
[10] R. Alcala, P. Ducange, F. Herrera, B. Lazzerini, F. Marcelloni, “A Multiobjec-tive Evolutionary Approach to Concurrently Learn Rule and Databases of Linguistic Fuzzy-Rule-Based Systems”, IEEE Transactions on Fuzzy Systems 17 (5) (2009) 1106–1122.
[11] T.A. Johansen, R. Babuska, “Multiobjective Identification of Takagi-Sugeno Fuzzy Models”, IEEE Transactions on Fuzzy Systems 11 (6) (2003) 847–860.
[12] H.. Wang, S. Kwong, Y. Jin, W. Wei, K.F. Man, “Agent-Based Evolutionary Approach for Interpretable Rule-Based Knowledge Extraction”, IEEE Trans-actions on Systems Man, and Cybernetics – Part C: Applications and Reviews 35 (2) (2005) 143–155.
[13] Y. Yoshinari, W. Pedrycz, K. Hirota, “Construction of Fuzzy Models Through Clustering Techniques”, Fuzzy Sets and Systems 54 (1993) 157–165.
[14] H. Ishibuchi, T. Yamamoto, T. Nakashima, “Fuzzy Data Mining: Effect of Fuzzy Discretization”, in: Proceedings of the First IEEE International Confe-rence on Data Mining, 2001, pp. 241–248.
[15] M. Setnes, R. Babuska, B. Verbruggen, “Rule-Based Modeling: Precision and Transparency”, IEEE Transactions on Systems, Man, and Cybernetics – Part C 28 (1) (1998) 165–169.
[16] Y. Jin, “Fuzzy Modeling of High-Dimensional Systems: Complexity Reduc-tion and Interpretability Improvement”, IEEE Transactions on Fuzzy Systems 8 (2) (2000) 212–221.
[17] D.E. Goldberg, “Genetic Algorithms in Search Optimization and Machine Learning”, Addison Wesley, Reading, MA, 1989.
[18] M. Sugeno, T. Yasukawa, “A Fuzzy-Logic-Based Approach to Qualitative Modeling”, IEEE Transactions on Fuzzy Systems 1 (1) (1993) 7–31.
[19] L.A. Zadeh, “A Fuzzy-Set-Theoretic Interpretation of Linguistic Hedges”, Journal of Cybernetics 2 (1972) 4–34.
[20] J. Yen, L. Wang, “Application of Statistical Information Criteria for Optimal Fuzzy Model Construction”, IEEE Transactions on Fuzzy Systems 6 (3) (1998) 362–372.
[21] J. Yen, L. Wang, “Simplifying Fuzzy Rule-Based Models Using Orthogonal Transformation Methods”, IEEE Transactions on Systems, Man, and Cyber-netics – Part B: Cybernetics 29 (1) (1999) 13–24.
[22] L. Wang, J. Yen, “Extracting Fuzzy Rules for System Modeling Using a Hybr-id of Genetic Algorithms and Kalman filter”, Fuzzy Sets and Systems 101 (1999) 353–362.
[23] M. Setnes, H. Roubos, “GA-Fuzzy Modeling and Classification: Complexity and Performance”, IEEE Transactions on Fuzzy Systems 8 (5) (2000) 509–522.
[24] D Agarwal, S..P. Tripathi, J..B. Singh., “TrFRA: A Trust Based Fuzzy Regres-sion Analysis”, Published in IRECOS Vol.5 N.6 November 2010, pp.668-670.
[25] D Agarwal, S..P. Tripathi, J..B. Singh, “Web Personalization of Indian e-Commerce Websites using Classification Methodologies”, Published in IJSCI Volume: 7 Issue: 6, November 2010, pp. 18-21.
Journal of Advances in Information Technology
ISSN 1798-2340
Volume 3, Number 4, November 2012
Contents
REGULAR PAPERS
Comparison of Segmentation Tools for Multiple Modalities in Medical Imaging Sonali Bhadoria, Preeti Aggarwal, C. G. Dethe, and Renu Vig
Comparison Four Different Probability Sampling Methods based on Differential Evolution Algorithm Qingbo Lu, Xueliang Zhang, Shuhua Wen, and Guosheng Lan
Nepenthes Honeypots Based Botnet Detection Sanjeev Kumar, Rakesh Sehga, Paramdeep Singh, and Ankit Chaudhary
Web based Geo-Spatial and Village Level Information Extraction System using FOSS Bheemashankar Gurupadayya Kodge and Prakash S. Hiremath
Design of Primary Screening Tool for Early Detection of Breast Cancer C. Naga Raju, C. Harikiran, and T. Siva Priya
Using Sequential Search Algorithm with Single level Discrete Wavelet Transform for Image Compression (SSA-W) Mohammed Mustafa Siddeq
E-Commerce: True Indian Picture Devendera Agarwal, R. P. Agarwal, J. B. Singh, and S. P. Tripathi
197
206
215
222
228
236
250
E-Commerce: True Indian Picture Devendera Agarwal
Research Scholar, Shobhit University, School of CE&IT, Meerut, India Email: [email protected]
R.P.Agarwal*, J.B.Singh* and S.P.Tripathi#
*Shobhit University, Meerut, India # Institute of Engineering Technology, Department of Computer Science, Lucknow, India
Email: [email protected], [email protected], [email protected]
Abstract—This paper gives an insight of e-commerce and highlights the present scenario of e-commerce in India. It presents the surfing pattern of Indian public to give the critical review on truth of various reports being published from time to time. It also critically analyses the e-commerce with major focus on B2C e-commerce which involves e-tailing.
Index Terms—e-Commerce, B2C, e-tailing, Indian Consumer, Trust
I. INTRODUCTION
India is a country with rich historical heritage, the second most populous country and the most populous democracy in the world. It has achieved multifaceted socio-economic progress during the last 63 years of its independence and has once again emerged on world scenario as one of largest economies.
India [23] primarily being country whose economy encompasses the traditional village farming, finally accepted computer as an ally not foe. Computer growth in India moved in leap and bounds after much hyped Y2K problem. Since then India have make its presence felt worldwide in software industry with companies like Infosys, TCS, Wipro etc. grew exponentially. According to the International Monetary Fund, India's nominal GDP stood at US$1.3 trillion, which makes it the eleventh-largest economy in the world, corresponding to a per capita income of US$1,000. If purchasing power parity (PPP) is taken into account, India's economy is the fourth largest in the world at US$3.6 trillion. The country ranks 142nd in nominal GDP per capita and 127th in GDP per capita at PPP. With an average annual GDP growth rate of 5.8% for the past two decades, India is one of the fastest growing economies in the world.
According to a 2011 PwC report [24], in terms of PPP, India's GDP will overtake that of Japan in 2011 and by 2045, India's GDP will surpass that of the United States. Additionally, over the next four decades, India's average annual economic growth rate is expected to stand at about 8% and therefore, it has the potential to be the world's fastest growing major economy over the period to 2050. India has large numbers of well-educated people skilled in English language; India is a major exporter of software services and software workers.
The Indian Information Technology industryaccounts for a 5.19% of the country's GDP and export earnings as of 2009, while providing employment to a significant number of its tertiary sector workforce. More than 2.3 million people are employed in the sector either directly or indirectly, making it one of the biggest job creators in India and a mainstay of the national economy.
Privatization of technical education also took place during this period and today India which churns out almost a million engineers every year. Public schools were not too far behind in imparting computer education as it was made compulsory from IIIrd standard itself. According to a study, titled “India Urban consumer segment nationwide study 2009-10”, surveyed 19,178 respondents across 82 cities by Intel and IMRB, it was reported that computer penetration in urban India doubled in last three years from 19 per cent to 38 per cent and now nearly 28 million households have the PC in their houses.
The PC purchases have been driven by better education opportunity, internet connectivity and ease of working from home. Multipurpose usage of PC for gaming, watching videos and listening to music has also kicked off the sales of PC.
Technical education in India is governed by All India Council of Technical Education (AICTE) [4], which makes it compulsory to maintain ratio of 1 computer for every 4 seats in an engineering institute. It is also compulsory to have dedicated internet connectivity with bandwidth of 1 Mbps or more. Under normal circumstances any institute running for over 4 years must have at least 300 PCs. Moreover more and more institutes are offering students laptops at subsidized rates. School children have also started to demand computers for practice at home.
This gives an insight to why computer sales have surged in last few years. Computer and Internet are two essential components for e-commerce. Their increase has certainly had a positive impact on e-commerce growth in India.
II. E-COMMERCE UNLEASHED
The electronic commerce concept was developed in the 70's even though electronic commerce under the
250 JOURNAL OF ADVANCES IN INFORMATION TECHNOLOGY, VOL. 3, NO. 4, NOVEMBER 2012
© 2012 ACADEMY PUBLISHERdoi:10.4304/jait.3.4.250-257
infant form of EDI or electronic data interchange has been existing since the late 60's with the invention of the first data networks [25].
Electronic commerce, commonly known as e-commerce or eCommerce, consists of the buying and selling of products or services over electronic systems such as the Internet and other computer networks. The amount of trade conducted electronically has grown extraordinarily with widespread Internet usage. The use of commerce is conducted in this way, spurring and drawing on innovations in electronic funds transfer, supply chain management, Internet marketing, online transaction processing, electronic data interchange (EDI), inventory management systems, and automated data collection systems. Modern electronic commerce typically uses the World Wide Web at least at some point in the transaction's lifecycle, although it can encompass a wider range of technologies such as e-mail as well.
A large percentage of electronic commerce is conducted entirely electronically for virtual items such as access to premium content on a website, but most electronic commerce involves the transportation of physical items in some way. Online retailers are sometimes known as e-tailers and online retail is sometimes known as e-tail. Almost all big retailers have electronic commerce presence on the World Wide Web.
Electronic commerce is generally considered to be the sales aspect of e-business. It also consists of the exchange of data to facilitate the financing and payment aspects of the business transactions.
Originally, electronic commerce was identified as the facilitation of commercial transactions electronically, using technology such as Electronic Data Interchange (EDI) and Electronic Funds Transfer (EFT). These were both introduced in the late 1970s, allowing businesses to send commercial documents like purchase orders or invoices electronically. The growth and acceptance of credit cards, automated teller machines (ATM) and telephone banking in the 1980s were also forms of electronic commerce.
In 1990, Tim Berners-Lee invented the WorldWideWeb web browser and transformed an academic telecommunication network into a worldwide everyman everyday communication system called internet/www. Commercial enterprise on the Internet was strictly prohibited until 1991. Although the Internet became popular worldwide around 1994 when the first internet online shopping started, it took about five years to introduce security protocols and DSL allowing continual connection to the Internet. By the end of 2000, many European and American business companies offered their services through the World Wide Web. Since then people began to associate a word "ecommerce" with the ability of purchasing various goods through the Internet using secure protocols and electronic payment services.
With the advent of the World Wide Web (WWW), electronic commerce and especially company-to-consumer electronic commerce, is based on public networks such as Internet. Their main characteristic being
that they are less expensive and widely accessible not only by corporations but also by the single individuals. There are many definitions of electronic commerce and much confusion there is about this term. For example Wigand [26] states that “Electronic commerce denotes the seamless application of information and communication technology from its point of origin to its endpoint along the entire value chain of business processes conducted electronically and designed to enable the accomplishment of a business goal. These processes may be partial or complete and may encompass business-to-business as well as business to consumer and consumer-to-business transactions”.
Zwass [27] defines electronic commerce as “The sharing of business information, maintaining business relationships, and conducting business transactions by means of telecommunications networks…Therefore as understood here, E-commerce includes the sell-buy relationships and transactions between companies, as well as the corporate processes that support the commerce within individual firms”.
A broader definition by Kalakota and Whinston [28] is: “E-commerce is associated with the buying and selling of information, products and services via computer networks today and in the future via any one of the myriad of networks that make up the Information Superhighway (I-way)”.
Internet also enables the marketers to easily reach the customers and promote their brands or products by offering vast product information and options. Electronic Commerce is the buying and selling of goods and services electronically by consumers or by companies via computerized transactions. E-Commerce has speeded up ordering, production, delivering, payment for goods and services by replacing manual and paper based business processes with electronic alternatives and by using information flow effectively in new and dynamic ways. At the same time, e-Commerce has reduced marketing, operational, production, and inventory costs in such a way that customer will also benefit indirectly.
Therefore, Internet is the technology [17, 18, 19] for e-Commerce as it offers easier ways to access companies and individuals at a very low cost in order to carry out day-to-day business transactions. Around the clock presence of companies on the Web gives competitive advantage to companies’ businesses.
However, since the Internet is publicly accessible, data can be more easily intercepted, which seriously undermines the security of online transactions, as well as the privacy and confidentiality of the commercial exchange.
Moreover, the legitimacy and the trustworthiness of online vendors cannot be guaranteed as adequately as on a private network, because there is no control as to who will enter the system and how parties will authenticate themselves. Since users will often have the choice between a large numbers of different business partners and since the cost of switching from one vendor to another is negligible, it is imperative that online vendors stand out by addressing not only users’ functional
JOURNAL OF ADVANCES IN INFORMATION TECHNOLOGY, VOL. 3, NO. 4, NOVEMBER 2012 251
© 2012 ACADEMY PUBLISHER
business needs, but also their concerns in terms of security, confidentiality and trustworthiness.
For private users to adopt e-commerce, it is imperative that the benefits of using the new commercial medium (e.g. convenience, decreased transaction costs) significantly outweigh potential risks. Indeed, the private user's freedom to select appropriate vendors tends to be correlated with greater concerns regarding financial risk, privacy and trust. This can be accounted for by the fact that private users are more directly involved in the commercial exchange, since they are using their own equipment, giving sensitive information about themselves as individuals, and spending their own money.
A. Types of E-Commerce The major different types of e-commerce are:
Business-To-Business (B2B) B2B e-commerce is simply defined as e-
commerce between companies. This is the type of e-commerce that deals with relationships between and among businesses. About 80% of e-commerce is of this type, and most experts predict that B2B ecommerce will continue to grow faster than the B2C segment.
Business-To-Consumer (B2C) Business-to-consumer e-commerce, or commerce
between companies and consumers, involves customers gathering information; purchasing physical goods (i.e., tangibles such as books or consumer products) or information goods (or goods of electronic material or digitized content, such as software, or e-books); and, for information goods, receiving products over an electronic network. It is the second largest and the earliest form of e-commerce. Its origins can be traced to online retailing (or e-tailing). Thus, the more common B2C business models are the online retailing companies such as Amazon.com, Drugstore.com, yahoo.com, rediff.com and indiatimes.com.
B2C e-commerce reduces transactions costs (particularly search costs) by increasing consumer access to information and allowing consumers to find the most competitive price for a product or service. B2C e-commerce also reduces market entry barriers since the cost of putting up and maintaining a Web site is much cheaper than installing a “brick-and-mortar” structure for a firm. In the case of information goods, B2C e-commerce is even more attractive because it saves firms from factoring in the additional cost of a physical distribution network. Moreover, for countries with a growing and robust Internet population, like India delivering information goods becomes increasingly feasible.
Business-To-Government (B2G) Business-to-government e-commerce or B2G is
generally defined as commerce between companies and the public sector. It refers to the use of the Internet for public procurement, licensing procedures, and other government-related operations.
A web-based purchasing policy increases the transparency of the procurement process (and reduces the risk of irregularities). To date, however, the size of the B2G ecommerce market as a component of total e-commerce is insignificant, as government e-procurement systems remain undeveloped.
Consumer-To-Consumer (C2C) Consumer-to-consumer e-commerce or C2C is
simply commerce between private individuals or consumers. This type of e-commerce is characterized by the growth of electronic marketplaces and online auctions, particularly in vertical industries where firms/businesses can bid for what they want from among multiple suppliers.
Mobile Commerce (m-commerce) M-commerce (mobile commerce) is the buying
and selling of goods and services through wireless technology i.e., handheld devices such as cellular telephones and personal digital assistants (PDAs).
As content delivery over wireless devices becomes faster, more secure, and scalable, some believe that m-commerce will surpass wire line e-commerce as the method of choice for digital commerce transactions. This may well be true for the Asia-Pacific where there are more mobile phone users than there are Internet users. This brief write-up with e-commerce discusses the
evolution of e-commerce and how it has become an almost necessity in our day to day life. Frequent developments in technology particularly 3G and 4G in mobile will only add to the speed of growth of e-commerce.
Our major emphasis will be on B2C, as this type e-commerce is e-retailing or more common e-tailing. It involves the process of billing the end consumer. In our view this is where the true test of e-commerce takes place.
E-commerce requires monetary transaction, one single step where user hesitates to complete transaction. He already has heard so many electronic frauds [14, 15, 16], computers being hacked, passwords stolen etc., on the contrary truth is only (0.03%) of B2C transaction are fraud. 87% of fraud comes from online auctions (C2C e-commerce). If this myth can be broken it will prove to be a big leap in e-commerce.
In B2B and B2G e-commerce one party is a business house while other being a business house or a government organization. Both of them are well aware of threats of e-frauds which include manipulation of data records, hacking into organization systems, manipulation computer programs, unauthorized transfer of funds, failure of an e-transactions etc. Both business houses and government organizations have cyber security cells to maintain their computer system networks with the help of firewalls, proxy servers, antivirus software, white-list authorized wireless connections etc. They also have facility of legal advice to handle cyber crimes. On other hand a consumer in very small fish in sea who wants cheapest products usually falls in trap knowingly or unknowingly.
252 JOURNAL OF ADVANCES IN INFORMATION TECHNOLOGY, VOL. 3, NO. 4, NOVEMBER 2012
© 2012 ACADEMY PUBLISHER
TABLE I. SURFING PATTERN IN INDIA
S. No Categories Total
1 Portals 36
2 Advertisement 16
3 Entertainment 14 4 Social Networking 13
5 Search Engines 7
6 Internet Service Providers 6 7 Banks 4 8 E-Commerce 3
9 Encyclopedia 1
Total 100
Figure 1. Sub-categories of Websites.
An e-commerce transaction requires a PC with internet connectivity and can be carried out from Home, Cybercafé or Office. At home we can assume PC to be safe, at office we have proper security of routers, ISA Server, Firewall and Antivirus so perhaps most secure place, while a cybercafé is perhaps the most susceptible place for a fraud.
It is essential to safeguard the interest of the consumer; it is he who will decide the fate of e-commerce in future. His trust has to be build; e-commerce will automatically grow with his trust and confidence. You can cheat him once, only to drive him away and not to trust e-commerce.
III. E-COMMERCE IN INDIA
Tall claims have been made about internet usage and e-commerce in India. Let’s not go by the amount, as in B2B the numbers of transactions are negligible but amount involved is huge. B2B has always been here in form of EDI, so why there is so much fuss. It is B2C and C2C e-commerce which constitute majority of transactions of comparatively small amount.
The study, titled ‘India Urban Consumer Segment Nationwide Study 2009-2010 surveyed 19,178 respondents across 82 cities by Intel and IMRB, it was reported that Computer penetration in urban India has doubled in the last three years from 19 per cent to 38 per cent and now nearly 28 million households have the PC in their houses. The study also noted that youth in the age group of 18-25 are able to play a significant role as facilitators during the actual purchase of the PC.
The PC purchases [22] have been driven by better education opportunity, internet connectivity and ease of working from home. Multipurpose usage of PC for Gaming, watching videos and listening to music has also kicked off the sales of PC.
Study also found that, more first-time buyers are buying notebooks as their first computer. In 2006, only a mere 17 per cent of first-time buyers wanted to go for notebooks, while in 2009, the percentage of non-owners who wanted a notebook instead of a desktop PC doubled to 31 per cent.
One thing is sure that internet usage in India is increasing in leap and bounds. Many surveys [6, 7, 8, 9, 10, 11, 12, 13] show same is the case with e-commerce. Lets not go by the amount, as in B2B the numbers of transactions are negligible but amount involved is huge. It has always been there in form of EDI, so why there is so much fuss. It is B2C or C2C e-commerce where we have many transactions of small amount.
In this section first we find out the surfing pattern of Indians, to get the answer to two primary questions.
a) Are internet users really interested in e-commerce?if answer to above question is yes than to find b) What we they buying on the internet and why?We found out the top 100 websites (hit ratio) [5] in
India and classified them; the result is shown in Table-1. Now let’s explain each category and try to assimilate
the necessary information on surfing behavior, and If possible, find out shopping pattern.
Portals: A web portal, presents information from diverse sources in a unified way. Apart from the
standard search engine feature, web portals offer other services such as e-mail, news, stock prices, information, databases and entertainment.
When further sub-categorized the picture came out as shown in figure-1. It is being observed that few portals are involved in e-commerce activities but keeping in mind Indian scenario, name of six websites is worth mentioning i.e. (Rediff India, Indiatimes, Ebay India & Sify) under General,Shaadi under Matrimonial and Makemytrip under Travel Sub-categories. Remaining other offer various other services and hence are not worth
mentioning.
JOURNAL OF ADVANCES IN INFORMATION TECHNOLOGY, VOL. 3, NO. 4, NOVEMBER 2012 253
© 2012 ACADEMY PUBLISHER
TABLE II. RANKING & %AGE OF INDIANS VISITING THESE WEBSITES
S.No Website IndianRank
World Rank
%age IndianAudience
1 Rediff India 9 145 89
2 Indiatimes 12 173 78
3 IRCTC 36 574 98
4 eBay India 52 854 95
5 Sify 64 833 83
6 Shaadi 90 978 76
7 Makemytrip 98 1335 96
(Source: as per data available from alexa.com)
TABLE III. CONTINGENCY TABLE FROM THE DATA OF TABLE II.
Red
iff
Indi
atim
es
IRC
TC
eBay
Indi
a
SIFY
Shaa
di
Mak
emyt
rip
Tota
l
IndianRank 9 12 36 52 64 90 98 361
WorldRank 145 173 574 854 833 978 1335 4892
Advertisement: Under these categories are those websites whom name one might have never heard of. They consist of those websites which generally pop on your screen when you visit other websites. They are basically advertisements on other websites and simply increasing their hit ratios Komli, Sulekha, Quikr India etc. are few of them. Entertainment: This category is yet another extension of portals, where whole emphasis is on entertainment only. It includes websites offering (Games, Music, Videos, Cricket scores etc.)Social Networking: It consist of latest in fashion sites meant for communication with friends, social causes etc. consist of common websites like Facebook, Twitter, Orkut, Bharatstudent etc.Search Engines: Perhaps one of most powerful tool on internet for actual working, no surprise, first two most popular websites being search engines Google and Google India, other are Bing, Ask, etc. Internet Service Providers (ISP): Again like advertisement category it comprises of websites which user actually doesn’t visit himself but their hitcounter automatically hits when we open some other websites like websites of Hit counter (StarCounter, Conduit, etc.) and Domain Names at any spelling/typing mistake in name (GoDaddy, DomainTools, etc.).Banks: Banks are financial institutions an essential requirement for carrying out monetary transactions in e-commerce. There are actually three banks with one bank having two domain names (HDFC, ICICI and SBI). HDFC and ICICI are largest private banks in India; they are also pioneers of Internet Banking in India. Any net savvy user will certain have an account in these banks. Third State Bank of India (SBI) is largest and the oldest bank in India also offers net banking facility. Most business organizations have account in it and again finding its name in this category is no surprise.
Presence of these names suggests that e-commerce is certainly present and making impact on India growth story. E-Commerce: Under these category we have placed those files that are strictly e-commerce (B2C) websites (irctc.co.in & amazon.com) and one being third party money transfer (paypal.com). Amazon is one of world most popular e-commerce websites, it is more likely that search engines direct user to Amazon rather user visiting this site for buying a product since payment is in Dollars & not in Rupees.PayPal is again facing problems with RBI
guidelines so its presence in top 100 may be due to B2B transactions or some other reason and not from B2C transactions. Website of Indian Railway Catering & Tourism Corporation is a classical example of e-commerce growth in India and we discuss it in next section in detail.
Encyclopedia: The single website of Wikipedia must attribute its presence to search engines, as it offers definition and history of each term being searched and it one of automatic choice for visiting the site.
In top 100 websites as per India’s choice and conditions seven websites clearly can be termed as involved in B2C as they are selling products or services to Indian public in their own currency, and four websites providing payment gateway to carry out these transactions.
IV. TOP E-COMMERCE WEBSITES IN INDIA
Our next step was to individually analyze top e-commerce websites and find out what they have to offer
and how much impact they make on India e-commerce growth. In the Table-II we have displayed these websites with their Indian Rank, World Rank and percentage of Indian audience visiting these websites.
Here Rediff India, Indiatimes and Sify can be grouped together in B2C category selling products of varied types Viz. mobiles, laptops, camera, watches etc., while eBay also deals in same but deals in C2C e-commerce.
We will be checking whether there exists a perfect combination between Indian Rank, World Rank and %age of Indian audience. To check this we use chi-square test [29, 30] of independence, considering the hypothesis as:
254 JOURNAL OF ADVANCES IN INFORMATION TECHNOLOGY, VOL. 3, NO. 4, NOVEMBER 2012
© 2012 ACADEMY PUBLISHER
TABLE V. COMPARATIVE PRICES OF VARIOUS PRODUCTS OFFERED BY
E-COMMERCE WEBSITES
S.No Category
E-commerce Websites (Price in Rs.)
Luc
know
Mar
ket
Mod
el N
o
Red
iff
Indi
a
eBay
Indi
a
Indi
atim
es
Sify
1 Mobile Nokia 7925 8899 9711 9099 10670
2 Laptops Compaq 32887 30888 33100 34999 33460
3 Camera Canon 6245 8267 8095 6590 8320
4 Watches FastTrack 1895 2240 2195 2200 2250
TABLE IV. CONTINGENCY TABLE FROM THE DATA OF TABLE III.
f0 fe D=f0-fe D*D D*D/fe
9 51.57 -42.57 1812.20 35.1407
12 51.57 -39.57 1565.78 30.3623
36 51.57 -15.57 242.42 4.7009
52 51.57 0.43 0.18 0.0036
64 51.57 12.43 154.50 2.9960
90 51.57 38.43 1476.86 28.6381
98 51.57 46.43 2155.74 41.8023
145 698.85 -553.85 306749.82 438.9351
173 698.85 -525.85 276518.22 395.6761
574 698.85 -124.85 15587.52 22.3045
854 698.85 155.15 24071.52 34.4445
833 698.85 134.15 17996.22 25.7512
978 698.85 279.15 77924.72 111.5042
1335 698.85 636.15 404686.82 579.0754
X2= 1751.3349
H0: PR1 = PR2H1: PR1 PR2
Where PR1 = Proportion of Indian Rank PR2 = Proportion of World Rank Combined proportion of Indian Rank
= 361/4892 = 0.07
We compare the observed value of x2 with critical values of x2 and apply the rules of Hypothesis:
x2observed < x2critical => Accept the Null Hypothesis and if
x2observed > x2critical => Reject the Null Hypothesis Now calculation the degree of freedom [31, 32] we get:
F = (r – 1) (c – 1) F = (2 – 1) (7 – 1) F = (1) (6) F = 6
at = 10% = 0.10 x2
critical / = 0.10
Since sample chi-square lies outside the acceptance region [33, 34] as shown in Figure-2 we reject the null
hypothesis i.e.: the ranking of websites in India and the World does not have any correlation with each other.
To find out, whether e-commerce in really useful other than convenience, time saving, etc., we visited all the B2C websites and found various similar products under different categories and also found their respective prices in surrounding area of living, making sure that we get the cheapest price and is also convenient to go and buy. To our surprise e-commerce also turned out to be the cheapest in all cases, with 7 days replacement guarantee and in most cases free postage and handling. The results of our finding are shown in Table V and Figure 3 represents the regression line showing relationship between Category & Prices.
JOURNAL OF ADVANCES IN INFORMATION TECHNOLOGY, VOL. 3, NO. 4, NOVEMBER 2012 255
© 2012 ACADEMY PUBLISHER
Two other websites of Indian Railway Catering & Tourism Corporation (IRCTC) and Makemytrip both are related to travel industry i.e. booking of air and rail tickets and reservation in hotels. According to news nearly 40% of booked tickets are sold online. According to a report by IAMI (Internet and Mobile Association in India) 75% of total e-commerce business comes from travel industry and rest everything in B2C and C2C category contribute only 25%. Indian Railways fourth largest in the world carries 20 million passengers daily and IRCTC being official website for booking tickets is no surprise biggest contributor to e-commerce in India.
In India railways tickets can either be booked at railway station or online yourself/ through agents. Most stations offer booking service from 8 AM to 8 PM and tickets sold are cheaper then purchased online. But the shear amount of time it takes to get ticket booked tells upon the patience of any person. You require half/ full day leave to get tickets booked. Thus we see no reason for people to prefer to book tickets online.
V. CONCLUSIONS & FUTURE SCOPE
We have seen that potential for growth of e-commerce in India is enormous. We have also seen that amount of interest that is there for travel industry is not seen in other services. Professional e-commerce websites [20, 21] are doing excellent job but what are the factors that are inhibiting users from purchasing online need to be ascertained.
The authors are working on the problem for last three years. They have tried to ascertain reasons in their papers [1, 2, 3] already communicated in referred journals.
In “Web Personalization of Indian e-Commerce Websites using Classification Methodologies” [2], authors have suggested how to identify fraud users by using Classification Methodologies using Bayesian Rules and generating cluster of users having fraudulent intentions.
In “TrFRA: A Trust Based Fuzzy Regression Analysis” [1], authors have focused on trust building factors in e-commerce websites. It focuses on fuzzy relationship between Trust and website related factors.
In “Trust Vs Complexity of E-commerce Sites” [3], authors have discussed how we need to tradeoff between Trust and Complexity so as not to drive away users from the website.
There are still various factors to be looked upon to cater to needs of the consumer who is the driving force for e-commerce his TRUST, CONVENIENCE, SECURITY are prime factors without which we will not be able to attain our goal.
ACKNOWLEDGMENT
The authors wish to thank management of Goel Group of Institutions, Lucknow for providing sponsorship in this endeavor.
REFERENCES
[1] Agarwal, Devendera, et. al., “TrFRA: A Trust Based Fuzzy Regression Analysis”, International Review on Computers and Software, Vol.5 N.6 November 2010, pp.668-670.
[2] Agarwal, Devendera, et. al., “Web Personalization of Indian e-Commerce Websites using Classification Methodologies”, International Journal of Computer Science Issues, Volume: 7 Issue: 6, November 2010, pp. 18-21.
[3] Agarwal, Devendera, et. al., “Trust Vs Complexity of E-commerce Sites”, to be published in International Journal of Scientific & Engineering, April 2012.
[4] All India Council of Technical Education (AICTE) website www.aicte-india.org
[5] ALEXA: The web Information Company website www.alexa.com
[6] Internet & Mobile Association of India (IAMAI) & Internet Market Research Bureau (IMRB), “I-CUBE 2008”, Report by IMRB International, India, 2008
[7] Internet & Mobile Association of India, “I-CUBE 2009-10”, Report by IMRB International, India, 2010.
[8] Internet & Mobile Association of India, “Cybecafe Users: E-commerce Activites”, Report 2005.
[9] Internet & Mobile Association of India & Internet Market Research Bureau, “Consumer E-commerce market In India”, Report 2006/07
[10] Internet & Mobile Association of India, “Annual Report 2006/07”, Report 2007.
[11] Internet & Mobile Association of India, “Annual Report 2007/08”, Report 2008.
[12] Internet & Mobile Association of India, “I-Cube 2007: Internet in India”, Report 2007.
[13] Internet & Mobile Association of India & Internet Market Research Bureau, “I-Cube 2008: Internet in India”, report 2008.
[14] Internet and Online Association of India (IOAI), “IOAI Survey: Ecommerce Security 2005”, Survey Report 2005
[15] KPMG, “India Fraud Survey Report 2006”, KPMG Forensic India.
[16] KPMG, “India Fraud Survey Report 2008”, KPMG Forensic India.
[17] PEW/Internet, “Online Shopping”, Survey Report 2008. [18] Humpery Jonh, et.al., “The Reality of E-commerce with
developing Countries”, Report 2003
256 JOURNAL OF ADVANCES IN INFORMATION TECHNOLOGY, VOL. 3, NO. 4, NOVEMBER 2012
© 2012 ACADEMY PUBLISHER
[19] Feldman Stuart, “The Changing face of E-commerce: Extending the Boundaries of the Possible”, IEEE Internet Computing, 2000.
[20] Yong-Mi Kim, Suliman Hawamdeh, “The Utulization of Web 2.0 Functionalities on E-commerce Websites”, Journal of Advances in Information Technology 2011, Vol2, No.4.
[21] K. Pragadeesh Kumar, Dr.N.Jaisankar, N.Mythili, “An Efficient Technique for Detection of Suspicious Malicious Web Site”, Journal of Advances in Information Technology 2011, Vol2, No.4.
[22] Rastogi, Rajiv, “INDIA: Country Report on E-Commerce Initiatives”, 2007.
[23] Wikipedia : The Free Encyclopedia (2011), “Introduction about India” Available: http://en.wikipedia.org/wiki/India
[24] Pwc Annual Report (2011), “Emerging Opportunities – Financial Services M&A in Asia 2011”, Available at http://www.pwc.com/gx/en/mergers-acquisitions-industry-trends/survey/index.jhtml
[25] Ravi Kalakota, Andrew B. Whinston (1997), “Electronic Commerce: A Manager’s Guide”, Addison Wesley.
[26] Wigand, R. T (1997), “Electronic Commerce: Definition, Theory and Context, The Information Society”, Vol 13, 1-16
[27] Zwass. V (1998), “Structure and Macro-Level Impacts of Electronic Commerce: From Technological Infrastructure to Electronic Market Places in Emerging Information Technologies” ed. Kenneth E. Kendall, Thousand Oaks, CA: Saga Publications.
[28] Ravi Kalakota , Andrew B. Whinston (1996), “Frontiers of electronic commerce”, Addison Wesley Longman Publishing Co., Inc., Redwood City, CA, 1996
[29] Greenwood, P.E., Nikulin, M.S. (1996), “A guide to chi-squared Testing”, Wiley, New York.
[30] Corder, G.W., Foreman, D.I. (2009), “Nonparametric Statistics for Non-Statisticians: A Step-by-Step Approach”, Wiley.
[31] Eisenhauer, J.G. (2008), “Degrees of Freedom”, Teaching Statistics, 30(3), 75–78
[32] Trevor Hastie, Robert Tibshirani, Jerome H. Friedman (2009), “The elements of statistical learning: data mining, inference, and prediction”, Springer 2nd ed., 746 p.
[33] E. L. Lehmann (1997), "Testing Statistical Hypotheses: The Story of a Book". Statistical Science
[34] Lehmann, E.L.; Joseph P. Romano (2005), “Testing Statistical Hypotheses” (3E ed.). New York: Springer.
Devendera Agarwal is research scholar at Shobhit University, Meerut pursuing his Ph.D. in Computer Science. He has done B.Tech. in Computer Science from Mangalore University, Mangalore in 1993 and M.Tech. from U.P. Technical University, Lucknow in 2006.
Presently he is working as Director at Goel Institute of Technology & Management, Lucknow since 2008. His
teaching areas are: Data Structures, Computer Graphics, Compiler Construction, Data Compression and Computer Programming. His areas of interest include Fuzzy Systems, E-commerce, Human computer interaction etc.
Dr. R. P. Agarwal is currently working as Vice Chancellor at Shobhit University, Meerut. He was former Vice Chancellor of Bundelkhand University, Jhansi. Prof. Agarwal has rich varied experience of teaching, research, development and administration. He has 41 years of teaching experience and has published more than 100 research papers in referred International and National Journals.
Dr. J.B. Singh is currently working Professor and Dean Students Welfare at Shobhit University, Meerut. His teaching areas are Bio-Statistics, Statistics, Mathematical Modeling, Statistical Computing, Crop Yield Estimation, Operation Research and Neural Network.
Dr. S.P. Tripathi is currently working as Assistant Professor in Department of Computer Science at Institute of Engineering Technology, Lucknow. He is M.Tech. from IIT, Delhi in 1985 and Ph.D. from Lucknow University in 2005. His teaching areas are Software Engineering, Database Management, Operating Systems, Data Mining and Computer Network.
JOURNAL OF ADVANCES IN INFORMATION TECHNOLOGY, VOL. 3, NO. 4, NOVEMBER 2012 257
© 2012 ACADEMY PUBLISHER