An Efficient Graph-based Relational Learning Algorithm
Zheng Li · 2008
Multi-relational data mining can be categorized into graph-based and logic-based according to their representation. We talk about the relationship between graph-based data mining and graph-based relational learning. An overview on different methods for graph-based data mining is given. We mainly discuss graph-based relational learning algorithm Subdue,including its advantage and disadvantage. To solves the disadvantages of Subdue,we propose ESubdue,which improve the subgraph isomorphism computation and reduces the times for subgraph isomorphism. Experimental results on both real and synthetic datasets indicate that the improved algorithm is much more efficient than the original one. Finally we conclude the paper and talk about the future work.