Evolutionary Rough K-Means Clustering

摘要

Rough K-means algorithm and its extensions have been useful in situations where clusters do not necessarily have crisp boundaries. Experimentation with the rough K-means algorithm has shown that it provides a reasonable set of lower and upper bounds for a given dataset. Evaluation of clustering obtained from rough K-means using various cluster validity measures has also been promising. However, rough K-means algorithm has not been explicitly shown to provide optimal rough clustering. This paper proposes an evolutionary rough K-means algorithm that minimizes a rough within-group-error. The proposal is different from previous Genetic Algorithms (GAs) based rough clustering, as it combines the efficiency of rough K-means algorithm with the optimization ability of GAs. The evolutionary rough K-means algorithm provides flexibility in terms of the optimization criterion. It can be used for optimizing rough clusters based on different criteria.