One Scan Sanitization of Informative Association Rules

Abstract

We propose here a one-scan sanitization algorithm to hide informative association rules. For a given predicting item, an informative association rule set by Jiuyong Li et. al, (2001) is the smallest association rule set that makes the same prediction as the entire association rule set by confidence priority. To hide association rules, previously proposed algorithms based on a priori approach require multiple scanning of database to calculate the supports of the large itemsets. In this work, we propose using a pattern-inversion tree to store related information so that only one scan of database is required. Numerical experiments show that the performance is more efficient than previous algorithms with similar side effects