Gustafson-Kessel-like Clustering Algorithm Based on Typicality Degrees
Marie‐Jeanne Lesot, Rudolf Kruse · WORLD SCIENTIFIC eBooks · 2008
Typicality degrees were defined in supervised learning as a tool to build characteristic representatives for data categories. In this paper, an extension of these typicality degrees to unsupervised learning is proposed to perform clustering. The proposed algorithm constitutes a Gustafson-Kessel variant and makes it possible to identify ellipsoidal clusters with robustness as regards outliers.