Hiding Sensitive Association Rules Using EMDSRRC

Marate Shashank S., Manjusha Yeola

Abstract: Association rules are generated to find out relation between item sets in database. So when it comes to large datasets, generating association rules becomes crucial. There are various techniques which are used to generate association rules such as Apriori algorithm and FP Growth algorithm. To find relation between item sets in database, association rule mining technologies are used. Many organizations uncover their information or data for mutual profits to find some useful information for some decision making purpose and improve their business. But this database may contain some secret data and which the organization doesn't want to uncover. In this paper, a heuristic based algorithm named EMDSRRC (Enhanced Modified Decrease Support of R. H. S. item of Rule Clusters) to hide the sensitive association rules with multiple items in consequent (R. H. S) and antecedent (L. H. S) is proposed. FP growth algorithm is used for generating rules and then selects items based on transactions to hide the sensitive information. We have proposed an algorithm EMDSRRC which uses FP growth algorithm to generate rules to overcome limitation in MDSRRC (Modified Decrease Support of R. H. S. item of Rule Clusters) which uses Apriori algorithm to generate rules.

Keywords: Association Rules, FP growth, Apriori, Sensitivity