Privacy-Preserving Data Mining (PPDM) Method for Horizontally Partitioned Data Mohamed A.Ouda ,Sameh A. Salem, Ihab A. Ali, and El-Sayed M.Saad Department of Communication and Computer, Faculty of Engineering Helwan University, Cairo - Egypt Abstract Due to the increase in sharing sensitive data through networks among businesses, governments and other parties, privacy preserving has become an important issue in data mining and knowledge discovery. Privacy concerns may prevent the parties from directly sharing the data and some types of information about the data. This paper proposes a solution for privately computing data mining classification algorithm for horizontally partitioned data without disclosing any information about the sources or the data. The proposed method (PPDM) combines the advantages of RSA public key cryptosystem and homomorphic encryption scheme. Experimental results show that the PPDM method is robust in terms of privacy, accuracy, and efficiency. 1. Introduction: Data mining is an important tool to extract patterns or knowledge from data [1]. Data mining technology can be used to mine frequent patterns, find associations, perform classification and prediction, etc. The data required for data mining process may be stored in a single database or in distributed resources. The classical approach for distributed resources is data warehouse. Fig. 1 shows a typical distributed data mining approach for building a data warehouse containing all the data. This requires the warehouse to be trusted and maintains the privacy of all parties. Since the warehouse knows the source of data, it learns site-specific information as well as global results. What if there is no such trusted authority? In a sense, this is a scaled-up version of the individual privacy problem; however it is an area where the Secure Multiparty Computation approach is more likely to be applicable. In this paper, RSA public key cryptosystem and homomorphic encryption are used to develop a reliable privacy-preserving data mining technique for horizontally partitioned data. The organization of the paper is as follows: Section 2 briefly describes the related work in the area. Section 3 gives background view about the techniques used as well as description of K nearest neighbor classifier as data mining technique. Section 4 presents the proposed algorithm satisfying privacy requirements. Section 5 presents experiments that are carried out to examine the performance of the proposed PPDM algorithm using three different real-world data sets. Section 6 presents a discussion of the experimental results. Section 7 concludes the paper and gives future directions for this research. Fig. 1: Data warehouse approach to mining distributed sources IJCSI International Journal of Computer Science Issues, Vol. 9, Issue 5, No 1, September 2012 ISSN (Online): 1694-0814 www.IJCSI.org 339 Copyright (c) 2012 International Journal of Computer Science Issues. All Rights Reserved.