A Novel Approach for Hiding Sensitive Association Rules using DPQR Strategy in Recommendation Systems

Mining association rules is considered to be a core topic of data mining. Discovering these associations is beneficial and is highly needed to the correct and appropriate decision made by decision makers in the different fields. Association rule Mining imposes threats to data sharing, since it may disclose patterns and various kinds of sensitive knowledge that are difficult to find. Such information must be protected against unauthorized access. In this paper, we are implementing DPQR strategy (data perturbation and query restriction) to hide the sensitive patterns. Experimental results showed that our proposed system can hide sensitive rules with multiple items in consequent (right hand side (R.H.S) ) and antecedent ( left hand side (L.H.S)) with efficient and faster performance compared to MDSRRC (Modified Decrease Support of R.H.S. items of Rule Cluster) with average improvement
96.22 % as well as generating accurate recommendations without revealing sensitive information.