A Hybrid Approach to Privacy Preserving in Association Rules Mining Abstract Nowadays, data mining is a useful, yet dangerous technology through which useful information and the relationships between items in a database are detected. Today, companies and users need to share information with others for their progress and they should somehow manage this information sharing for preserving sensitive information. Privacy preserving in data mining was introduced for managing information sharing. This paper presents a hybrid algorithm with distortion technique with both support-based and confidence-based approaches for privacy preserving. The proposed algorithm tries to maintain useful association rules and hide sensitive rules from the perspective of the database owner. It also has no limit on the number of items in the left-hand side and the right-hand side of rules. This paper also compares the proposed algorithm with MDSRRC algorithm and 1.b algorithm. The proposed algorithm has less lost rules compared with the MDSRRC and 1.b algorithms and its CPU usage is less then. Keywords: privacy preserving, hiding sensitive rules, helpful association rules. 1. Introduction The data mining technique aims to detect useful rules and relationships of database items for which standard algorithms are used such as Apriori and Eclat. The detected rules are divided into two groups: sensitive rules and non-sensitive rules. Sensitive rules are those rules the database owner is trying to hide using privacy preserving algorithms in data mining, and non-sensitive rules are useful rules the database owner wants to share with others. Of course, doing anything has its own costs. The cost the database owner pays for hiding sensitive rules is the loss of non-sensitive rules plus other costs we called them side effects due to hiding sensitive rules. These side effects include loss of non-sensitive rules, creating ghost rules, hiding failure, dissimilarity, runtime, etc. The algorithms presented in the context of privacy preserving have tried to reduce these side effects. The proposed algorithm aims to reduce lost rules and reduce hiding failure to zero. Generally there are two approaches for hiding sensitive rules [1]: support-based approach and confidence-based approach. The support-based approach aims to reduce support of the sensitive rule by reducing support of one of element sets composing the sensitive rule. The confidence-based approach aims is reducing confidence of the sensitive rule through increasing support of the consequent of the sensitive rule. This paper uses both approaches for hiding sensitive rules. The proposed algorithm selects either approach to sanitize by calculating the support of left-hand side and right-hand side elements of the sensitive rule. In this algorithm, selecting the item and the transaction to sanitize has a large impact on reducing side effects. This paper compares the proposed algorithm with MDSRRC algorithm on the Chess dataset. The article is organized as follows. Section 2 describes the framework of association rule. Section 3 reviews the related works. Section 4 explains the proposed algorithm, terms and steps. Section 5 compares and evaluates the proposed algorithm with MDSRRC and 1.b. Finally, conclusions are presented in Section 6. 2. Framework of association rules Rule extraction in data mining is performed by the level of support and confidence of the rule. The issue of extracting association rule was introduced by [2]. Suppose I={i 1 ,i 2 ,…,i m }is a set of elements and the database D={T 1 ,…,T n }is a set of transactions. Each transaction T∊D contains a subset of I. The general framework of association rules is X⟶Y. If X and Y are subsets of I and if xy, then X is called the antecedent or LHS of the rule and Y is called the consequent or RHS. The support of the rule X⟶Y is defined by calculating the ratio of simultaneous repetition frequencies of X and Y in transactions to the total number of transactions in the database. The support of the rule is calculated by Eq. (1). ( 1 )      The confidence of the rule X⟶Y is defined by calculating the ratio of simultaneous repetition frequencies of X and Y in ACSIJ Advances in Computer Science: an International Journal, Vol. 3, Issue 6, No.12 , November 2014 ISSN : 2322-5157 www.ACSIJ.org 69 Copyright (c) 2014 Advances in Computer Science: an International Journal. All Rights Reserved. Narges Jamshidian Ghalehsefidi 1 , Mohammad Naderi Dehkordi 2 1 jamshidian_n@sco.iaun.ac.ir, 2 naderi@iaun.ac.ir 1,2 Department of Computer Engineering, Najafabad Branch, Islamic Azad University, Najafabad, Iran