CLUSTERING WITH EVOLUTION STRATEGIES

The applicability of evolution strategies (ESs), population based stoc hastic optimization techniques, to optimize clustering objective funct ions is explored. Clustering objective functions are categorized into centroid and non-centroid type of functions. Optimization of the centr oid type of objective functions is accomplished by formulating them as functions of real-valued parameters using ESs. Both hard and fuzzy cl ustering objective functions are considered in this study. Applicabili ty of ESs to discrete optimization problems is extended to optimize th e non-centroid type of objective functions. As ESs are amenable to par allelization, a parallel model (master/slave model) is described in th e context of the clustering problem. Results obtained for selected dat a sets substantiate the utility of ESs in clustering.