Graph-Based Hierarchical Conceptual Clustering

Hierarchical conceptual clustering has been proven to be a useful data mining technique. Graph-based representation of structural information has been shown to be successful in knowledge discovery. The Subdue substructure discovery system provides the advantages of both approaches. In this paper we present Subdue and focus on its clustering capabilities. We use two examples to illustrate the validity of the approach both in structured and unstructured domains, as well as compare Subdue to an earlier clustering algorithm.

• Type: Conference
• Year: 2000
• Where: FLAIRS
• Authors: Istvan Jonyer, Lawrence B. Holder, Diane J. Cook
• Keywords: Artificial Intelligence, Earlier Clustering Algorithm, FLAIRS 2000, Hierarchical Conceptual Clustering, Subdue Substructure Discovery