Upload
ijafrc
View
220
Download
0
Embed Size (px)
DESCRIPTION
Data mining is used for various ways like government sectors, private sectors eg. Banking,Research center. Data mining is the classification process. Now a days, due to the growth ofInternet various privacy issues arise. In classification problem different security model will beprepared. Now a days, due to the growth of cloud computing, user can outsource their data oncloud in encrypted form and can apply data mining tasks on cloud. In data mining privacypreservingclassification technique are not applicable. So, in our paper we are solving the privacypreservingclassification problem over encrypted data. We are using k-NN classifier techniqueover encrypted data in cloud. The proposed system provides-Confidentiality of data, Privacy ofuser input query, Hides data access patterns
Citation preview
International Journal of Advance Foundation and Research in Computer (IJAFRC)
Volume 2, Issue 10, October - 2015. ISSN 2348 4853, Impact Factor 1.317
1 | 2015, IJAFRC All Rights Reserved www.ijafrc.org
A Novel Privacy- Preserving K-NN Classification Protocol
Over Encrypted Data Mr. Kumud Ranjan Kumar, Mr. Ankush Prasun, Mr. Nityanand, Ms. Anjali Jadhav
Imperial College of Engineering and Research, Pune,Maharashtra India
[email protected], [email protected]
A B S T R A C T
Data mining is used for various ways like government sectors, private sectors eg. Banking,
Research center. Data mining is the classification process. Now a days, due to the growth of
Internet various privacy issues arise. In classification problem different security model will be
prepared. Now a days, due to the growth of cloud computing, user can outsource their data on
cloud in encrypted form and can apply data mining tasks on cloud. In data mining privacy-
preserving classification technique are not applicable. So, in our paper we are solving the privacy-
preserving classification problem over encrypted data. We are using k-NN classifier technique
over encrypted data in cloud. The proposed system provides-Confidentiality of data, Privacy of
user input query, Hides data access patterns
Index Terms : k-NN Classifier, AES algorithm, Normalization.
I. INTRODUCTION
The DMED (Data mining over encrypted data) problem is not solve in Data mining. In this paper, we are
proposed to solve the DMED problem on cloud. It also protects the data of user when record is apart of a
Data mining process.
Cloud can be used for getting the useful and sensitive (Private data) information from the actual data by
keeping the eyes on pattern used by the user even if the data is in the encrypted for. The Privacy-
preserving issue is solved in cloud, it provides-
(1) Confidentiality of the encrypted data.
(2) Confidentiality of a users query record.
(3) Hiding data access patterns.
The proposed system is focuses on executing k-nearest neighbor classification Technique over encrypted
data in the cloud computing environment.
II. Explanation:
In this paper, we are using k-nearest neighbor classification technique for solving the DMED problem.
Normalization technique and AES algorithm is using for encryption of data. The proposed system is
providing privacy-Preserving classification over encrypted data on Cloud.
A. Normalization Technique:
International Journal of Advance Foundation and Research in Computer (IJAFRC)
Volume 2, Issue 10, October - 2015. ISSN 2348 4853, Impact Factor 1.317
2 | 2015, IJAFRC All Rights Reserved www.ijafrc.org
Normalization is the process of managing tables and column for minimizing redundant data.
Normalization may have the effect of duplicating data within the database and often results in the
creation of additional tables.
B. K-nearest neighbor classification technique:
K nearest neighbors is a simple algorithm that stores all available cases and classifies new cases based on
a similarity measure (e.g., distance functions).
Algorithm:
A case is classified by a majority vote of its neighbours, with the case being assigned to the class most
common amongst its K nearest neighbours measured by a distance function. If K = 1, then the case is
simply assigned to the class of its nearest neighbour.
Example:
Consider the following data concerning credit default. Age and Loan are two numerical variables
(predictors) and Default is the target.
International Journal of Advance Foundation and Research in Computer (IJAFRC)
Volume 2, Issue 10, October - 2015. ISSN 2348 4853, Impact Factor 1.317
3 | 2015, IJAFRC All Rights Reserved www.ijafrc.org
We can now use the training set to classify an unknown case (Age=48 and Loan=$142,000) using Euclidean distance.
If K=1 then the nearest neighbor is the last case in the training set with Default=Y.
D = Sqrt[(48-33)^2 + (142000-150000)^2] = 8000.01 >> Default=Y
With K=3, there are two Default=Y and one Default=N out of three closest neighbors. The prediction for
The unknown case is again Default=Y.
C. AES algorithm:
AES (acronym of Advanced Encryption Standard) is a symmetric encryption algorithm. The block to be
encrypted is just a sequence of 128 bits. AES works with byte quantities so we first convert the 128 bits
into 16 bytes.
International Journal of Advance Foundation and Research in Computer (IJAFRC)
Volume 2, Issue 10, October - 2015. ISSN 2348 4853, Impact Factor 1.317
4 | 2015, IJAFRC All Rights Reserved www.ijafrc.org
AES Block Diagram:
The encryption process uses a set of specially derived keys called round keys. These are applied, along
with other operations, on an array of data that holds exactly one block of data. The data to be encrypted.
This array we call the state array.
AES Steps of encryption for a 128-bit block:
1. Derive the set of round keys from the cipher key.
2. Initialize the state array with the block data (plaintext).
3. Add the initial round key to the starting state array.
4. Perform nine rounds of state manipulation.
5. Perform the tenth and final round of state manipulation.
6. Copy the final state array out as the encrypted data (ciphertext).
III. Architecture Diagram:
International Journal of Advance Foundation and Research in Computer (IJAFRC)
Volume 2, Issue 10, October - 2015. ISSN 2348 4853, Impact Factor 1.317
5 | 2015, IJAFRC All Rights Reserved www.ijafrc.org
In this system, User inputs data to server i.e. user upload data on server after that we use data mining
process for data part ion we calculate mean and deviation after that normalize data. Server do the
encryption of that data then when user demand data server decrypt that data after that we apply the K-
NN algorithm and we improve the classification of data.
IV. Flow Diagram
All the data will be first stored in to data set until it gets loaded. Partition will be done on the basis of
their individual characteristic of the data. Mean and standard deviation will take place after
Normalization. Encryption will be carried out on Normalized data. Encrypted data can be decrypted by
The authorized user and the related data can be searched by using K-nearest neighbor algorithm. After
that some accurate data will find.
International Journal of Advance Foundation and Research in Computer (IJAFRC)
Volume 2, Issue 10, October - 2015. ISSN 2348 4853, Impact Factor 1.317
6 | 2015, IJAFRC All Rights Reserved www.ijafrc.org
V. CONCLUSION
The proposed system provides the user privacy. It also solves the DMED problem using k-NN
classification technique over encrypted data n cloud. A novel privacy- Preserving k-NN classification
protocol over encrypted data in cloud
VI. REFERENCES
[1] S. De Capitani di Vimercati, S. Foresti, and P. Samarati,Managing and accessing data in the cloud:
Privacy risks andapproaches, in CRiSIS, pp. 1 9, 2012.
[2] Y. Elmehdwi, B. K. Samanthula, and W. Jiang, Secure k-nearest neighbour Query over encrypted
data in outsource environments, in IEEE ICDE, pp. 664675, 2014.
[3] L. Xiong, S. Chitti, and L. Liu, K nearest neighbor Classification across multiple private
databases, in pp. 840841, ACM, 2006.
International Journal of Advance Foundation and Research in Computer (IJAFRC)
Volume 2, Issue 10, October - 2015. ISSN 2348 4853, Impact Factor 1.317
7 | 2015, IJAFRC All Rights Reserved www.ijafrc.org
[4] X. Xiao, F. Li, and B. Yao, Secure nearest neighbor revisited, in IEEE ICDE, pp. 733744, 2013.
[5] R. Agrawal and R. Srikant, Privacy-preserving data mining, in ACM Sigmod Record, vol. 29,pp.
439 450, ACM, 2000.
[6] Y. Lindell and B. Pinkas, Privacy preserving data mining, in Advances in Cryptology (CRYPTO),
pp. 3654, Springer, 2000.
[7] P. Zhang, Y. Tong, S. Tang, and D. Yang, Privacy preserving nave bayes classification, ADMA, pp.
744752, 2005.
[8] Y. Qi and M. J. Atallah, Efficient privacy-preserving k-nearest neighbor search, in IEEE ICDCS, pp.
311319, 2008.
[9] A. Evfimievski, R. Srikant, R. Agrawal, and J. Gehrke, Privacy preserving mining of association
rules, Information Systems, vol. 29, no. 4, pp. 343364, 2004.
[10] D. Bogdanov, S. Laur, and J. Willemson, Sharemind: A framework for fast privacy-preserving
computations, in ESORICS, pp. 192206, Springer, 2008