Frequent Subgraph Mining on BigData

Nowadays, graph mining is important role in various domains. The application of graph mining are chemical field, social networks, traffic analysis etc. The frequent subgraph mining is that extract the frequent subgraphs from graph dataset. In this paper, the frequent graph mining implemented on Spark. The Spark can take less time comparatively Hadoop MapReduce task. It contain three phases. They are Preprocessing phase, Mining phase and Prediction phase. The Preprocessing is the first phase. The Mining phase, mine the frequent subgraphs. Then prediction phase, predict the nodes, which is infected in future. The prediction phase predict the node based on mining phase. This application helps to hospital field.