METHOD FOR MINING FREQUENT SUBGRAPH USING MAPREDUCE

摘要

The present invention relates to a method for mining a frequent subgraph in a large-size graph database. More specifically, the frequent subgraph can be rapidly mined in the large-size graph database by a distributed process on the large-size graph database by using a Mapreduce method; and a computation amount and time for generating a candidate subgraph can be reduced by generating a restored graph for each graph identifier from the frequent subgraph and then generating only the candidate subgraph capable of edge extension from the restored graph.