Graph theoretic based algorithm for mining frequent patterns

The primary goals of any frequent pattern mining algorithm are to reduce the number of candidates generated and tested as well as number of scan of database required and scan the database as small as possible. In this paper, we focus on reducing database scans and avoiding candidate generation. To achieve this objective a graph theoretic algorithm has been developed. The whole database is compressed by converting into pattern base in the form of a directed graph which is stored in the form of an Adjacency Matrix. This Adjacency Matrix is very small as compared to the size of database. This frequent pattern mining is done by performing operation on adjacency matrix of directed graph. The prominent feature of this method is it requires only single scan of the database for finding frequent patterns.