Impact factor based data sanitization in association rule mining

Abstract

Abstract Data sanitization process is employed to market the sharing of transactional databases among organizations and businesses, and alleviates concerns for people and organizations regarding the disclosure of sensitive patterns sanitization process converts the source database into a released database so that unauthorized person cannot identify the sensitive patterns and so data confidentiality is maintained using association rule mining method. This process strongly relies on the minimizing the impact of knowledge sanitization on the info utility by minimizing the amount of lost patterns within the sort of non– sensitive patterns which are not mined from sanitized database. This study proposes a knowledge sanitization algorithm to cover sensitive patterns within the sort of frequent item sets from the database while controlling the impact of sanitization on the data utility using estimation of impact factor of every modification on non-sensitive item sets. In some applications like market basket analysis, Association Rule Mining (ARM) has recently gained more attention in businesses where the regularities within the customer purchasing behavior are found. On the other hand, these discovered patterns may pose a threat to the privacy of data holder; therefore, these patterns should be hidden before data sharing in such a way that the adversaries cannot discover the regularities in customer purchasing behavior.