Fuzzy Clustering to Identify Clusters at Different Levels of Fuzziness: An Evolutionary Multiobjective Optimization Approach

Fuzzy clustering methods identify naturally occurring clusters in a dataset, where the extent to which different are overlapped can differ. Most have parameter fix level of fuzziness. However, appropriate fuzziness depends on application at hand. This paper presents an entropy c-means (ECM), method fuzzy that simultaneously optimizes two contradictory objective functions, resulting creation with levels allows ECM degrees overlap. functions using multiobjective optimization methods, nondominated sorting genetic algorithm II (NSGA-II) and evolutionary based decomposition (MOEA/D). We also propose select suitable tradeoff from Pareto front. Experiments challenging synthetic datasets as well real-world show leads better cluster detection compared conventional previously used for clustering.