Graph theoretical clustering based on limited neighbourhood sets

Abstract

A method for clustering data according to a visual model of clusters is proposed. The method uses either of two graphs which are defined according to relative distance and based on the Gabriel graph and the relative neighbourhood graph respectively. The method is locally sensitive, hierarchic and based on the concept of limited neighbourhood sets. Clusters that are either disjoint or homogeneous and separable by sharp changes in point density may be detected. © 1982.