CloseGraph: mining closed frequent graph patterns

摘要: Recent research on pattern discovery has progressed form mining frequent itemsets and sequences to structured patterns including trees, lattices, graphs. As a general data structure, graph can model complicated relations among with wide applications in bioinformatics, Web exploration, etc. However, large challenging due the presence of an exponential number subgraphs. Instead all subgraphs, we propose mine closed patterns. A g is database if there exists no proper supergraph that same support as g. algorithm, CloseGraph, developed by exploring several interesting pruning methods. Our performance study shows CloseGraph not only dramatically reduces unnecessary subgraphs be generated but also substantially increases efficiency mining, especially