A Fuzzy Variant of an Evolutionary Algorithm for Clustering

A fuzzy version of an evolutionary algorithm for clustering (EAC) proposed in previous work is introduced. This algorithm uses a fuzzy cluster validity criterion and a fuzzy local search algorithm instead of their hard counterparts employed by EAC. It is shown by means of theoretical complexity analyses that this algorithm can be more efficient than systematic (i.e. repetitive) approaches when the number of clusters is unknown. An illustrative example with computational experiments and statistical analyses is also presented.