GDClust: A Graph-Based Document Clustering Technique

This paper introduces a new technique of document clustering based on frequent senses. The proposed system, GDClust (graph-based document clustering) works with frequent senses rather than frequent keywords used in traditional text mining techniques. GDClust presents text documents as hierarchical document-graphs and utilizes an apriori paradigm to find the frequent subgraphs, which reflect frequent senses. Discovered frequent subgraphs are then utilized to generate sense-based document clusters. We propose a novel multilevel Gaussian minimum support approach for candidate subgraph generation. GDClust utilizes English language ontology to construct document-graphs and exploits graph-based data mining technique for sense discovery and clustering. It is an automated system and requires minimal human interaction for the clustering purpose.