Main Article Content
Graphs become increasingly important in modeling complicated structures such as chemical compounds, social networks, bioinformatics and protein networks. With the increasing demand on the analysis of large amounts of structured data, where applications include indexing, clustering and classification. Graph mining has become an active and important theme in data mining. Existing graph mining algorithms have achieved great success by exploiting various properties in the frequent patterns. Due to large size of graphs, finding the frequent patterns become complex and consume more time and also more enumeration cost. The solution we propose for this problem is, reduce the original size of graph. For each graph we build summary of it and mine the reduced graph. This can be achieved by SUMMARIZE-MINE-FRAMEWORK. Summarize-mine is effective in cutting down the size of original graphs, thus reduction in enumeration cost. However compression might lose patterns at the same time we address this issue by repeat the process for multiple rounds where the patterns miss from all rounds. Experiments on family network data show that SUMMARIZE-MINE is efficient, which finds frequent patterns easily and quickly.