Privacy preserving data mining (PPDM) has been a new research area in the past two decades. In fact, the aim of PPDM algorithms is to modify data in the dataset so that sensitive data and confidential knowledge, even after data mining operation be kept confidential. Association rule hiding is one of the main techniques of PPDM aiming to avoid extracting some rules that are recognized as sensitive rules. Most of the work which has been done in the area of association rule hiding are limited to binary data, however many real world datasets include quantitative data too. In this work a new methods is proposed to hide sensitive quantitative association rules which is based on convex optimization technique. In most of the previous methods, there were uniform changes in the values of all items and also in all of them the correlation between related items were not considered. By considering these two issues, fewer changes are made in the real and fuzzy dataset. In our proposed method, we make suitable changes in the value of each item and the relations between related items are defined as constraints in the optimization problem. In all existing methods at this field, association rules were extracted from 2-large itemsets. However our method can be extended for any kind of association rules. The performance of the proposed algorithm is measured in the term of percentage of hiding of sensitive rules, side effects and changes occurred in the fuzzy and real datasets. The results showed, most sensitive rules were made hidden with the proposed method and the number of lost and ghost rules and changes in the fuzzy and real datasets have been significantly reduced. Keyword : Data Mining, Association Rule Hiding, Convex Optimization
