Algorithms for Clustering Terms in Document Set Based on Fuzzy Neighborhoods
Satoshi Miyamoto, E. Kataoka · 2005
This paper describes similarity measures between two terms in a document set using the concept of a fuzzy neighborhood and algorithms for term clustering. Theoretical properties of neighborhood and similarity measures are studied. Agglomerative hierarchical as well as fuzzy/crisp c-means clustering algorithms are proposed. Examples of agglomerative and c-means clustering are given