HEART DISEASE PREDICTION SYSTEM USING NAIVE BAYES · advantage fr om knowing a patient's medicinal history, and machine learning isn't compelling unless extensive databases of data

HEART DISEASE PREDICTION SYSTEM USING

NAIVE BAYES Arun.R, UG Scholar, Saveetha School of Engineering, SIMATS. [email protected]

Mrs. N.Deepa, Asst Professor, Saveetha School of Engineering, SIMATS. [email protected]

ABSTRACT

Cloud-helped portable observing, which applies the across the board versatile interchanges and distributed

computing innovations to give input choice help, has been considered as a lobbyist way to deal with

enhancing the nature of social insurance benefit while bringing down the human services cost. Tragically, it

additionally represents a genuine hazard on both customer's security and licensed innovation of checking

specialist organizations, which could dishearten the wide selection of Health innovation. This venture is to

address this essential issue and outline a cloud-helped security saving versatile wellbeing checking

framework to ensure the protection of the included gatherings and their information. Besides, the

outsourcing decoding method and a recently proposed key private intermediary re-encryption are adjusted to

move the computational many-sided quality of the included gatherings to the cloud without trading off

customers' protection and specialist organizations' licensed innovation. At long last, our security and

execution examination shows the viability of our proposed plan. Demonstrating secure and execution

investigation shows the viability in Cloud Computing condition. Human services is a basic piece of life.

Lamentably, the relentlessly maturing populace and the related ascent in unending disease issue setting huge

strain on present day social insurance frameworks and the interest for assets from healing facility beds to

specialists and medical attendants is to a great degree high. Obviously, an answer is required to diminish the

weight on social insurance frameworks while proceeding to give brilliant care to in danger patients.

Restorative data acquired from patients must be put away safely for proceeded with utilize. Specialist’s

advantage from knowing a patient's medicinal history, and machine learning isn't compelling unless

extensive databases of data are accessible to it. In view of the writing, distributed storage is the most feasible

strategy for putting away information. Nonetheless, giving openness to medicinal services experts without

bargaining security is a key concern.

Keyword: Data mining Naive Bayes, heart disease, prediction,AES.

1 INTRODUCTION

In this quick moving world individuals need to

carry on with an extremely sumptuous life so they

work like a machine so as to acquire parcel of

cash and carry on with an agreeable life

subsequently in this race they neglect to deal with

themselves, in view of this there nourishment

propensities change their whole way of life

change, in this sort of way of life they are more

strained they have circulatory strain, sugar at an

exceptionally youthful age and they don't give

enough rest for themselves and eat what they get

and they even don't make a fuss over the nature of

the sustenance if debilitated the go for their own

medicine because of all these little carelessness it

prompts a noteworthy risk that is the coronary

illness. Heart is the most fundamental organ in

human body if that organ gets influenced then it

likewise influences the other key parts of the

body. In this manner it is vital for individuals to

go for a coronary illness analysis [1]. Because of

this individuals go to social insurance experts

however the forecast made by them isn't 100%

exact. Quality administration infers diagnosing

patients accurately and directing medications that

are viable. Poor clinical choices can prompt tragic

results which are in this manner unsuitable. They

International Journal of Pure and Applied MathematicsVolume 119 No. 16 2018, 3053-3065ISSN: 1314-3395 (on-line version)url: http://www.acadpubl.eu/hub/Special Issue http://www.acadpubl.eu/hub/

3053

can accomplish these outcomes by utilizing

suitable PC based data and additionally choice

emotionally supportive networks [2]. The human

services industry gathers tremendous measures of

social insurance information which, lamentably,

are not "mined" to find shrouded data for

powerful basic leadership. Disclosure of shrouded

examples and connections frequently goes

unexploited. Propelled information mining

systems can help cure this circumstance. This

exploration has built up a model Heart Disease

Prediction System (HDPS) utilizing information

mining methods, in particular, Decision Trees,

Naïve Bayes and Neural Network. Results

demonstrate that every method has its novel

quality in understanding the destinations of the

characterized mining objectives [3]. HDPS can

answer complex imagine a scenario where

inquiries which conventional choice emotionally

supportive networks can't. Utilizing medicinal

profiles, for example, age, sex, pulse and glucose

it can anticipate the probability of patients getting

a coronary illness. It empowers noteworthy

information, e.g. designs, connections between

medicinal variables identified with coronary

illness, to be built up. HDPS is Web-based, easy

to understand, versatile, dependable and

expandable. Here the extent of the venture is that

mix of clinical choice help with PC based patient

records could diminish medicinal blunders,

upgrade tolerant security, diminish undesirable

practice variety, and enhance understanding result

proposal is promising as information

demonstrating and examination devices, e.g.,

information mining, can possibly produce a

learning rich condition which can help to

altogether enhance the nature of clinical choices

.The fundamental goal of this exploration is to

build up a model Heart Disease Expectation

System (HDPS) utilizing three information mining

displaying strategies, to be specific, Decision

Trees, Naive Bayes and Neural Network [4]. So it

gives successful medicines, it likewise diminishes

treatment costs and furthermore upgrades

perception and simplicity of translation. With

enormous learning and exact information in that

field. Substantial organizations put vigorously in

this sort of movement to help concentrate

consideration on conceivable occasions and

dangers that are included. Such work unites all

accessible past and current information, as a

premise on which to create sensible assumptions

about what's to come.

We propose a Naive Bayes is a simple

technique for constructing classifiers:they

calculate the probability of each category

for a given sample, and then output the

category with the highest one. The way

they get these probabilities is by using

Bayes’ Theorem, which describes the

probability of a feature, based on prior

knowledge of conditions that might be

related to that feature.

We apply Advanced Encryption Standard

(AES) is an encryption algorithm for

securing sensitive but unclassified

material. AES is based on a design

principle known as a substitution

permutation network, a combination of

both substitution and permutation, and is

fast in both software and hardware. AES

performed well on a wide variety of

hardware, from 8-bit smart cards to high-

performance computers.

The relationship of this paper is according to the

accompanying: The accompanying region (2)

discusses the related work. In fragment (3), the

proposed demonstrate is displayed trailed by

appraisal and results examination in section (4).

Finally, we complete the paper and inspect future

course in fragment (5).

2 RELATED WORK

As of late, there has been a creating

eagerness for using sharp home advancements for

perceiving human development plans for

prosperity checking applications. The principal

objective is to take in inhabitants' behavioural

characteristics as an approach to manage

appreciate and anticipate their activities that could

indicate therapeutic issues.

In this section, we overview existing work

in the composition, which use insightful homes

International Journal of Pure and Applied Mathematics Special Issue

3054

data to separate customers' direct. Remote

versatile wellbeing checking has just been

perceived as not just a potential. It likewise

represents a genuine hazard on customers'

security. Licensed innovation of checking

specialist organizations additionally getting

challenges in innovation.

This paper is to address this vital issue and

plan a cloud-helped security. It saving portable

wellbeing observing framework to secure the

protection of the included gatherings and their

information. The outsourcing decoding method

and a recently proposed key private intermediary

reencryption are adjusted. It uses to move the

computational many-sided quality of the included

gatherings to the cloud without trading off

customer’s protection and specialist organizations

licensed innovation.

Adequately secure the protection of

customers and the licensed innovation of

wellbeing specialist organizations. Based the

encryption system, it secure the customer's

protection. To decrease the decoding intricacy, we

apply as of late proposed unscrambling

outsourcing. With access to online wellbeing data,

patients are progressively occupied with their own

care and research; this incorporates for all intents

and purposes each age, sexual orientation, and

race fragment and those with ceaseless wellbeing

conditions like cardiovascular malady and

diabetes mellitus.[5]Third, with the help of

another presidential activity, the idea of exactness

prescription is developing: Precision

pharmaceutical energizes anticipation and

treatment procedures to be custom-made to an

individual patient's genotypic and phenotypic

information (counting information gathered by

portable devices).[6] Finally, gadgets and

health‐monitoring "wearable's" that can be

coordinated into everyday life offer guarantee for

early determination and wellbeing advancement.

Cardiovascular ailment (CVD) is the main

source of death in the United States as well as on

the planet [7]; consequently, it is one of the

biggest health‐related regions in which mHealth

could propel medicinal services for patients,

suppliers, payers, and policymakers. A few

modifiable way of life practices are related with

the danger of creating CVD, and control of these

practices can altogether bring down a man's

lifetime CVD chance. With an incipient

confirmation base, numerous mHealth

intercessions have the guarantee of advancing

fruitful and supported behavioural change. Thusly,

inside the field of CVD, we delineated particular

open doors for mHealth; potential difficulties to

the improvement and selection of arrangements;

and a system for creating protected, viable, and

evidence‐based mHealth answers for CVD.

Furthermore, our experiments are conducted using

a much larger dataset than existing techniques

between the proposed study and existing work.

3 PROPOSED MODEL

It could be utilized to give better access to human

services to those living in provincial territories, or

to empower elderly individuals to live freely at

home for more. Basically, it can enhance access to

social insurance assets while decreasing strain on

human services frameworks, and can give

individuals better control over their own

wellbeing constantly. This is especially valid for

cardiovascular sickness, its decent variety of

indications and abundance of powerful

intercessions are conceivably mind boggling and

befuddling. Evaluating the potential populace

advantage of a wellbeing mediation requires


3055

thought of numerous components including

infection commonness and populace attributes,

viability and cost. Medicinal data got from

patients must be put away safely for proceeded

with utilize. Specialist’s advantage from knowing

a patient's medicinal history, and machine

learning isn't successful unless substantial

databases of data are accessible to it. In view of

the writing, distributed storage is the most

reasonable strategy for putting away information.

Giving availability to medicinal services experts

without trading off security is a key concern. Give

treatment designs and diagnostics, and offer

proposals to medicinal services experts that are

particular to singular patients.

3.1 DATA PREPARATION

The dataset used in this study is a collection of

smart meters data from Hungarian Institute of

Cardiology. Budapest: Andras Janosi, M.D. This

database contains 76 attributes, but all published

experiments refer to using a subset of 14 of them.

In the first stage of the cleaning process we

developed customized procedures to remove

noises from the data and prepare it for mining.

After cleaning and preparation, the dataset is

reduced to 20 million. Additionally, we developed

a synthetic dataset for preliminary evaluation of

the model, having over 1.2 million records. In

tables (1)we show the example of the resulting

ready to mine source data format with 14

attributes. The attributes values of min and max

values are given in the table.

TABLE (1)

3.2 IMPLEMENTATION OF NAIVE BAYES

CLASSIFIER

A. Classifier

A classifier is a process of mapping from a

(discrete or continuous) feature space X to a

discrete set of labels Y. Here we are dealing about

learning classifiers, and learning classifiers are

divided into supervised and unsupervised learning

classifiers [8]. The applications of classifiers are

wide ranging. They find use in medicine, finance,

mobile phones, computer vision (face recognition,

target tracking), voice recognition, data mining

and uncountable other areas. An example is a

classifier that accepts a person's details, such as

age, marital status, home address and medical

history and aclassifies the person with respect to


3056

the conditions of the project. Table (1) shows the

accuracy of the Naive Bayes classification.

B. Naive Bayes

In probability theory, Bayes' theorem (often called

Bayes' law after Thomas Bayes) relates the

conditional and marginal probabilities of two

random events. It is often used to compute

posteriorprobabilities given observations [9]. For

example, a patient may be observed to have

certain symptoms. Bayes' theorem can be used to

compute the probability that a proposed diagnosis

is correct, given that observation. A naive Bayes

classifier is a term dealing with a simple

probabilistic classification based on applying

Bayes' theorem. In simple terms, a naïve Bayes

classifier assumes that the presence (or absence)

of a particular feature of a class is unrelated to the

presence (or absence) of any other feature. For

example, a fruit may be considered to be an apple

if it is red, round, and about 4" in diameter. Even

though these features depend on the existence of

the other features, a naive Bayes classifier

considers all of these properties to independently

contribute to the probability that this fruit is an

apple. Depending on the precise nature of the

probability model, naive Bayes classifiers can be

trained very efficiently in a supervised learning

setting [10].

Naive Bayes classifiers often work much better in

many complex real-world situations than one

might expect. Here independent variables are

considered for the purpose of prediction or

occurrence of the event. In spite of their naive

design and apparently oversimplified assumptions,

naive Bayes classifiersoften work much better in

many complex real world situations than one

might expect. Recently,careful analysis of the

Bayesian classification problem has shown that

there are some theoretical reasons for the

apparently unreasonable efficacy of naive Bayes

classifiers []. An advantage of the naive Bayes

classifier is that it requires a small amount of

training data to estimate the parameters (means

and variances of the variables) necessary for

classification. Because independent variables are

assumed, only the variances of the variables for

each class need to be determined and not the

entire covariance matrix.[11]

C. Bayesian Theorem:

Given training data X, posterior probability of a

hypothesis H, P(H|X), follows the Bayes theorem

P(H|X)=P(X|H)P(H)/P(X)

Algorithm:

The Naive Bayes algorithm is based on Bayesian

theorem as given by equation[12]

Steps in algorithm are as follows:

1.Each data sam

ple is represented by an n dimensional feature

vector, X = (x1, x2….. xn), depicting n

measurements made on the sample from n

attributes, respectively A1, A2, An.

2. Suppose that there are m classes, C1,

C2……Cm. Given an unknown data sample, X

(i.e., having no class label), the classifier will

predict that X belongs to the class having the

highest posterior probability, conditioned if and

only if: P(Ci/X)>P(Cj/X) for all 1< = j< = m and

j!= i Thus we maximize P(Ci|X). The class Ci for

which P(Ci|X) is maximized is called the

maximum posteriori hypothesis. By Bayes

theorem,

3. As P(X) is constant for all classes, only

P(X|Ci)P(Ci) need be maximized. If the class

prior probabilities are not known, then it is

commonly assumed that the classes are equally

likely, i.e. P(C1) = P(C2) = …..= P(Cm), and we

would therefore maximize P(X|Ci). Otherwise, we

maximize P(X|Ci)P(Ci). Note that the class prior

probabilities may be estimated by P(Ci) = si/s ,

where Si is the number of training samples of

class Ci, and s is the total number of training

samples. on X. That is, the naive probability

assigns an unknown sample X to the class Ci(2).


3057

D. Bayesian interpretation

The Bayesian Classifier is capable of calculating

the most probable output depending on the input.

It is possible to add new raw data at runtime and

have a better probabilistic classifier. A naive

Bayes classifier assumes that the presence (or

absence) of a particular feature of a class is

unrelated to the presence (or absence) of any other

feature, given the class variable. For example, a

fruit may be considered to be an apple if it is red,

round, and about 4" in diameter. Even if these

features depend on each other or upon the

existence of other features, a naive Bayes

classifier considers all of these properties to

independently contribute to the probability that

this fruit is an apple.[13]

E. USES

Using medical profiles such as age, sex, blood

pressure and blood sugar, chest pain, ECG graph

etc.

1. It can predict the likelihood of patients getting

a heart disease. [14]

2. It will be implemented in PYTHON as an

application which takes medical test’s parameter

as an input.

3. It can be used as a training tool to train nurses

and medical students to diagnose patients with

heart disease.

F.IMPLEMENTATION OF BAYESIAN

CLASSIFICATION

The Naïve Bayes Classifier technique is

mainly applicable when the dimensionality of

the inputs is high. Despite its simplicity, Naive

Bayes can often outperform more

sophisticated classification methods. Naïve

Bayes model recognizes the characteristics of

patients with heart disease. It shows the

probability of each input attribute for the

predictable state. [15]

Predictable attribute:-

1. Diagnosis (value 0: <50% diameter

narrowing (no heart disease); value 1:

>50% diameter narrowing (has heart

disease)). It shows that in fig (2).

3.3 Advanced Encryption Standard (AES)

AES is a 128-bit block cypher, and the scheme

has already been optimized to run on small, lower-

power devices. It is an accepted industry standard.

ABE focuses on allowing access to multiple

authorized parties, much like the other schemes

discussed in this section. The three types of

encryption were compared in depth, and it was

found that there is no perfect solution for

encryption. AES is the only scheme that is simple

and low-latency enough for wearable devices, but

ABE is the best scheme for enabling multiple

parties to access private data. AES overall

diagram shown in fig (1). [16],[17].

Security remains a key issue in cloud-based

systems. In a healthcare environment, it is

essential that a patient’s health information is

readily accessible to authorized parties including


3058

doctors, nurses, specialists, and emergency

services. It is also essential that the patient’s

sensitive health data is kept private. If malicious

attacks revealed the patient’s health data, it could

have many negative ramifications for the patient,

including exposing them to identity theft or

making it difficult for them to obtain insurance.

Worse still, if the malicious attacker altered a

patient’s health record, it could have detrimental

effects on the patient’s health.

Access control policies and data encryption are

two means of securing cloud-centric healthcare

systems. An access control policy specifies who is

authorized access to the patient’s health data, and

how much access they are allowed. It would also

implement an authentication mechanism (e.g.

password, facial recognition, etc.) that verifies the

identity of the party attempting to access the data.

Meanwhile, data encryption provides security for

the data whilst in data storage. Strong data

encryption would prevent an attacker from

reading sensitive health information, even if they

did gain access to the database.

Some research has been conducted into

developing security mechanisms robust enough

for healthcare applications. In [18], a sophisticated

access control scheme named “SafeProtect” is

proposed, focusing on giving patients control over

their information. The patient creates a policy that

allows specific healthcare providers to access their

health record, and can enforce limitations. The

patient’s data is encrypted and stored in cloud

storage. If a healthcare provider wants to access

the patient’s health record, they must enter their

credentials. Credentials are checked for validity

before data is decrypted for the authorized

healthcare provider. If the healthcare provider has

been assigned limited access by the patient, then

the policy control application will monitor their

behavior. If illegal actions are performed, such as

the user pressing the Ctrl + C shortcut for a party

that is not allowed to copy, then the action will be

blocked and the patient will be notified that the

healthcare provider tried to perform an illegal

action. This is an intelligent mechanism to and

could easily be expanded to look for attempts to

print or other copy-based actions such as taking a

screenshot using the “Prt Scr” button. A

significant advantage of the Safe Protect scheme

is that if the policy changes, keys do not need to

be regenerated; healthcare providers’ credentials

can simply be added to or removed from the

policy. The authors identify that they have not

protected against all possible means of copying

and distributing healthcare information, making

this the main area in which future improvement

can be made. Additional monitoring of keyboard

shortcuts and ports for illegal actions would help

increase security. [19] Another potential

improvement is immediately revoking access to

the patient’s data if the healthcare provider

performs an illegal action. However, these

improvements are minor and relatively easy to

make. Overall, Safe Protect is a sophisticated

scheme that adequately protects patient’s data

from being accessed or used in an unauthorized

manner, and the mechanisms implemented could

easily be expanded to protect against more

actions. [20]

Inner Workings of a Round

The algorithm begins with an Add round key

stage followed by 9 rounds of four stages and a

tenth round of three stages. This applies for both

encryption and decryption with the exception that

each stage of a round the decryption algorithm is

the inverse of its counterpart in the encryption

algorithm. The four stages are as follows:

1. Substitute bytes

2. Shift rows

3. Mix Columns

4. Add Round Key

The tenth round simply leaves out the Mix

Columns stage. The first nine rounds of the

decryption algorithm consist of the following:

1. Inverse Shift rows

2. Inverse Substitute bytes

3. Inverse Add Round Key

4. Inverse Mix Columns

Again, the tenth round simply leaves out the

Inverse Mix Columns stage.[21] Each of these

stages will now be considered in more detail.


3059

FIG (1)

A)Substitute Bytes

This stage (known as Sub Bytes) is simply a table

lookup using a 16×16 matrix of byte values called

an s-box. This matrix consists of all the possible

combinations of an 8 bit sequence (28 = 16 × 16 =

256). However, the s-box is not just a random

permutation of these values and there is a well-

defined method for creating the s-box tables. The

designers of Rijndael showed how this was done

unlike the s-boxes in DES for which no rationale

was given. We will not be too concerned here how

the s-boxes are made up and can simply take them

as table lookups. Again the matrix that gets

operated upon throughout the encryption is known

as state. [22]

We will be concerned with how this matrix is

effected in each round. Fig(2)

Fig (2): Substitute Bytes Stage of the AES

algorithm

B) Shift Row Transformation

This stage (known as Shift Rows) is shown in

fig(3). This is a simple permutation an nothing

more. It works as follow:

• The first row of state is not altered.

• The second row is shifted 1 bytes to the left in a

circular manner.

• The third row is shifted 2 bytes to the left in a

circular manner.

• The fourth row is shifted 3 bytes to the left in a

circular manner. [23]

Fig(3): ShiftRows stage.

C) Mix Column Transformation

This stage (known as MixColumn) is basically a

substitution but it makes use of arithmetic of GF

(28). Each column is operated on individually.

Each byte of a column is mapped into a new value

that is a function of all four bytes in the column.

The transformation can be determined by the

following matrix multiplication on state (see

figure 4): [24]


3060

Fig (4):Mixcolumns stage.

D) Add Round Key Transformation

In this stage (known as Add Round Key) the 128

bits of state are bitwise XORed with the 128 bits

of the round key. [25] The operation is viewed as

a column wise operation between the 4 bytes of a

state column and one word of the round key. This

transformation is as simple as possible which

helps in efficiency but it also effects every bit of

state.

AES 9mplementation

To ensure the customer's security, we apply the

unknown AES in therapeutic analytic fanning

programs. [26] To diminish the unscrambling

intricacy because of the utilization of AES, we

apply as of late proposed decoding outsourcing

with security assurance to move customer's

blending calculation to the cloud

server.(see.fig(5).

Fig (5) AES

IMPLEMENTATION.

Token generation

To produce the private key for the quality vector,

a customer initially figures the character portrayal

set of every component in and conveys all the

personality portrayal sets to TA. [27] At that point

TA runs the on every personality in the character

set and conveys all the individual private keys to

the customer. [28]

Cipher text retrieval:

The cloud is required to produce the figure

writings for customers by running the Re

Encryption calculation. Each keep running of Re

Encryption calculation costs the cloud precisely

two blending calculations. [28] For every

customer, the cloud needs to play out those

Computations. [30] The subsequent open key

figure messages alongside the first symmetric key

figure writings constitute the Cipher content sets

for the customer. [29]

4 EVALUATION AND RESULTS

For the assessment of the proposed demonstrate,

we played out our analyses utilizing the dataset

alongside the manufactured dataset to investigate

middle of the road and last outcomes. This

database contains 76 properties, yet all distributed

analyses allude to utilizing a subset of 14 of them.

This is one of the biggest datasets having

estimated a large portion of a billion records.

Vitality utilization estimation was directed at

apparatus level utilizing module singular machine

screens (IAMs). The hidden framework for the

proposed show is created in Python, and the

information is put away in MySQL and Mongo

DB databases on a Ubuntu 14.04 LTS 64-bit

framework. The primary goal of the examinations

is to distinguish the apparatus utilization as a sign

of human action examples and utilize the

expectation model to estimate the short and long

haul exercises inside the house. For a medicinal

services application, this implies our model can be

utilized to nourish instruments, for example,

dynamic observing, ready age, wellbeing profiling

and so on.

5 RESULT ANALYSIS AND DESCRIPTION:


3061

Restorative finding is considered as a critical yet

many-sided undertaking that should be completed

unequivocally and effectively. The

computerization of the same would be

exceptionally helpful. Clinical choices are

frequently made in view of specialist's instinct and

experience instead of on the information rich

information covered up in the database. This

training prompts undesirable inclinations,

blunders and extreme medicinal costs which

influences the nature of administration gave to

patients. Information mining can possibly create a

learning rich condition which can help to

essentially enhance the nature of clinical

choices.Results and analysis is done on Cleveland

dataset. Results are shown in the form of pie

charts, bar charts. Table 1 shows the accuracy

obtained by changing the number of instances in

the training dataset.

6 Conclusion

In this paper, we design a cloud-assisted

privacy preserving mobile health monitoring

system, called CAM, which can effectively

protect the privacy of clients and the intellectual

property of mHealth service providers. To protect

the clients’ privacy, we apply the anonymous

Boneh–Franklin identity-based encryption (IBE)

in medical diagnostic branching programs. To

reduce the decryption complexity due to the use of

IBE, we apply recently proposed decryption

outsourcing with privacy protection to shift

clients’ pairing computation to the cloud server.

To protect mHeath service providers’ programs,

we expand the branching program tree by using

the random permutation and randomize the

decision thresholds used at the decision branching

nodes. Finally, to enable resource-constrained

small companies to participate in mHealth

business, our CAM design helps them to shift the

computational burden to the cloud by applying

newly developed key private proxy reencryption

technique. Our CAM has been shown to achieve

the design objective.

7 Future Work

We elaborated on both objective and

subjective elements of information concerning the

patient, acquired via a wearable monitoring

system and self-reporting by the patient

himself/herself, respectively. An additional

objective of this study was to enable controlled

sharing of the obtained information with


3062

caregivers/family members of the patient by

taking advantage of the social networking

paradigm. Our future work involves the further

development of methodologies for handling

contextual data, behavioral monitoring based on

user-to-system interactions, and appropriate

methods for the collaborative filtering of

information and discovery of patterns. In

particular, we aim to investigate ways for

inferring activity information rather than requiring

from the user to provide this information.

REFERENCE: [1]. Blake, C.L., Mertz, C.J.: “UCI Machine Learning

Databases”,

http://mlearn.ics.uci.edu/databases/heartdisease/, 2004.

[2]. Giudici, P.: “Applied Data Mining: Statistical Methods

for Business and Industry”, New York: John Wiley, 2003.

[3]. Wu, R., Peters, W., Morgan, M.W.: “The Next

Generation Clinical Decision Support: Linking Evidence to

Best Practice”, Journal Healthcare Information

Management. 16(4), 50-55, 2002.

[4]. Intelligent Heart Disease Prediction System Using Data

Mining Techniques-Sellappan Palaniappan, Rafiah Awang

978-1-4244-1968-5/08/ ©2008 IEEE

[5]. Fox S. Food safety, drug safety, and pregnancy

information are among eight new topics included in our

survey. Pew Research Center web

site. http://www.pewinternet.org/2011/02/01/health-topics-

2/. Updated February 1, 2011. Accessed December 3, 2015.

[6]. Collins FS, Varmus H. A new initiative on precision

medicine. N Engl J Med. 2015;372:793–795.[PubMed]

[7]. World Health Organization . The Global Burden of

Disease: 2004 Update. Geneva, Switzerland: World Health

Organization; 2008. Available

at: http://www.who.int/healthinfo/global_burden_disease/G

BD_report_2004update_full.pdf. Accessed May 8, 2016.

[8]. Chapman, P., Clinton, J., Kerber, R. Khabeza, T.,

Reinartz, T., Shearer, C., Wirth, R.: “CRISP-DM 1.0: Step

by step data mining guide”, SPSS, 1-78, 2000.

[9]. Fayyad, U: “Data Mining and Knowledge Discovery in

Databases: Implications for scientific databases”, Proc. of

the 9th Int. Conf. on Scientific and Statistical Database

Management, Olympia, Washington, USA, 2-11, 1997.

[10]. Intelligent Heart Disease Prediction System Using

Data Mining Techniques-Sellappan Palaniappan, Rafiah

Awang 978-1-4244-1968-5/08/ ©2008 IEEE

[11]. Ho, T. J.: “Data Mining and Data Warehousing”,

Prentice Hall, 2005.

[12]. Wu, R., Peters, W., Morgan, M.W.: “The Next

Generation Clinical Decision Support: Linking Evidence to

Best Practice”, Journal Healthcare Information

Management. 16(4), 50-55, 2002.

[13]. Chapman, P., Clinton, J., Kerber, R. Khabeza, T.,

Reinartz, T., Shearer, C., Wirth, R.: “CRISP-DM 1.0:

Step by step data mining guide”, SPSS, 1-78, 2000.

[14]. Obenshain, M.K: “Application of Data Mining

Techniques to Healthcare Data”, Infection Control and

Hospital Epidemiology, 25(8), 690–695, 2004.

[15]. SellappanPalaniappan, RafiahAwang, Intelligent Heart

Disease Prediction System Using Data Mining Techniques,

©2008

IEEE.

[16]. I. Olaronke, O. Oluwaseun, "Big data in healthcare:

Prospects challenges and resolutions", Proc. Future

Technol. Conf. (FTC), pp. 1152-1157, Dec. 2016.

[17]. J. Zhou, Z. Cao, X. Dong, A. V. Vasilakos, "Security

and privacy for cloud-based IoT: Challenges", IEEE

Commun. Mag., vol. 55, no. 1, pp. 26-33, Jan. 2017.

[18]. D. Thilakanathan, S. Chen, S. Nepal, R. Calvo,

"SafeProtect: Controlled data sharing with user-defined

policies in cloud-based collaborative environment", IEEE

Trans. Emerg. Topics Comput., vol. 4, no. 2, pp. 301-315,

Apr./Jun. 2016.

[19]. J. Pacheco, C. Tunc, P. Satam, S. Hariri, "Secure and

resilient cloud services for enhanced living

environments", IEEE Cloud Comput., vol. 3, no. 6, pp. 44-

52, Nov./Dec. 2016.

[20]. S.-D. Bao, M. Chen, G.-Z. Yang, "A method of signal

scrambling to secure data storage for healthcare

applications", IEEE J. Biomed. Health Informat., vol. 21,

no. 6, pp. 1487-1494, Nov. 2017.

[21]. Parikh, C., Parimal Patel, M.S.: Ph.D. Performance

Evaluation of AES Algorithm onVarious Development

Platforms.

[22]M. S. Hossain, “Cloud-supported cyber-

physical localization framework for patients

monitoring,” IEEE Systems Journal, vol. 11, no. 1, pp. 118–

127, March 2017.

[23]M. S. Hossain, G. Muhammad, W. Abdul, B. Song, and

B. Gupta,“Cloud-assisted secure video transmission and


3063

sharing framework for smart cities,” Elsevier, Future

Generation Computer Systems Journal, April 2017.

[24]J. Liao, L. Stankovic, and V. Stankovic, “Detecting

household activity patterns from smart meter data,”

in Intelligent Environments (IE), 2014 International

Conference on, 6 2014, pp. 71–78.

[25]A. Yassine, A. A. N. Shirehjini, and S. Shirmohammadi,

“Smart meters big data: Game theoretic model for fair data

sharing in deregulated smart grids,” IEEE Access, vol. 3,

pp. 2743–2754, 2015.

[26]A. Yassine and S. Shirmohammadi, “Measuring users’

privacy payoff using intelligent agents,” in 2009 IEEE

International Con- ference on Computational Intelligence

for Measurement Systems and Applications, May 2009,

pp. 169–174.

[27] “A business privacy model for virtual

communities,” In- derscience Publishers International

journal of web based communities, vol. 5, 2009.

[28]Y. C. Chen, H. C. Hung, B. Y. Chiang, S. Y. Peng, and

P. J. Chen, “Incrementally mining usage correlations among

appliances in smart homes,” in Network-Based Information

Systems (NBiS), 2015 18th International Conference on, 9

2015, pp. 273–279.

[29]K. Jack and K. William, “The UK-DALE dataset,

domestic appliance-level electricity demand and whole-

house demand from five UK homes,” Scientific Data, vol. 2,

no. 150007, 2015.

[30] J. Clement, J. Ploennigs, and K.Kabitzsch, Detecting

Activities of Daily Living with Smart Meters. Springer,

Germany, 11 2014, ch. Advance Technology and Societal

Change, pp. 143–160. [Online]. Available:

https://link.springer.com/chapter/10.1007/ 978-3-642-

37988-8 10


3064

3065

3066

Documents

HEART DISEASE PREDICTION SYSTEM USING NAIVE BAYES · advantage fr om knowing a patient's medicinal history, and machine learning isn't compelling unless extensive databases of data