Graph Based Frequent Item Set Generation in Association Rule Mining

This research proposes a new technique that uses a reconstruction strategy for categorical datasets to preserve the privacy of the association rules. In order to identify all frequent item association sets and determine sets comprising frequent items based on the association sets, the suggested technique takes advantage of the two-item permutation set to create an association part. The enhanced method scans the database only once, addressing a flaw in the conventional approach that scans the database twice and so increasing scanning costs. The database and association sets can be mined using the two-item permutation set, which has the advantages of operating more quickly, requiring less memory, and being simple to maintain.