Frequent subgraph mining algorithms on weighted graphs

This thesis describes research work undertaken in the field of graph-based knowledge discovery (or graph mining). The objective of the research is to investigate the benefits that the concept of weighted frequent subgraph mining can offer in the context of the graph model based classification. Weighted subgraphs are graphs where some of the vertexes/edges are considered to be more significant than others. How to discover frequent sub-structures with different strengths is the main issue to be resolved in this thesis. The main approach to addressing this issue is to integrate weight constraints into the frequent subgraph mining process. It is suggested that the utilization of weighted frequent subgraph mining generates more discriminate and significant subgraphs, which will have application in, for example, the classification and clustering of graph data.