Heuristic Optimized Algorithm for Hiding Sensitive Association Rules for Privacy Preserving Data Mining Gehad Ahmed Soltan Abd El Aleem 1 , Laila Abd-Ellatif Abd-Elmegid 2 , and Ahmed Sharaf Eldin Ahmed 3 1 Information Systems Department, Faculty of Computers and Information, Helwan University, Cairo, Egypt Eng.gehadahmed2013@yahoo.comm 2 Information Systems Department, Faculty of Computers and Information, Helwan University, Cairo, Egypt drlaila_mohamed@yahoo.com 3 Information Systems Department, Faculty of Computers and Information, Helwan University, Cairo, Egypt profase2000@yahoo.com Abstract-Association rule mining is a data mining technique that recognizes the frequent patterns and specifies the associative rules which predict occurrence of a pattern based on occurrence of other patterns. Association rules may be used to expose the confidentiality and privacy of data. Privacy Preserving Data Mining (PPDM) is the specialized research area used to handle privacy threats in data mining. One of the techniques of PPDM is Association Rule Hiding (ARH). In this paper we proposed a heuristic optimized algorithm for hiding sensitive association rules, (HOAHSAR). The proposed algorithm overcomes the main problem of similar existing techniques. The experimental study proves the efficiency of the proposed algorithm on different types of data sets. The efficiency is measured by three factors; number of lost rules, number of required modifications, and the execution time. A comparison between the proposed algorithm and the two famous FHSAR and HSARWI algorithms shows the superiority of the proposed algorithm. KEYWORDS Association Rules (AR), Association Rules Hiding (ARH), Data Mining (DM), Heuristic Approach, Privacy Preserving Data Mining (PPDM), Sanitized Data, Sensitive Association Rules (SAR) I. INTRODUCTION The main objective of data mining is the extracting hidden and implicit knowledge from data. It gives new insights of the data through various techniques including clustering, classifications, and association rules. The discovered knowledge is beneficial in strategic decision making. Privacy is a critical issue that should be handled efficiently in many applications that deal with sensitive critical data like security, medical, and financial applications. The data of such applications should be accessed only by authorized users. However, ensuring the protection of sensitive data by access restriction only is not a sufficient method. The results of data mining may threaten the privacy as they can be used to infer sensitive information. Privacy preserving data mining PPDM is the area dedicated to develop solutions for protecting sensitive information from unnecessary or unlawful disclosure by different data mining techniques. Association rules mining aims to discover the correlations among items Association rule hiding is one of the privacy preserving techniques that aims to hide sensitive association rules so that no sensitive information can be International Journal of Computer Science and Information Security (IJCSIS), Vol. 16, No. 11, November 2018 53 https://sites.google.com/site/ijcsis/ ISSN 1947-5500