The Evolutionary Algorithm of Fuzzy Weighting Exponent Based on Subset Measuring

As there still isn´t any theoretical foundation or effective evaluation for the definition of fuzzy weighting exponent at present, a kind of evolutionary algorithm of fuzzy weighting exponent based on subset measuring is presented during the application of Fuzzy C-Means (FCM). Firstly, a clustering validity function is defined based on subset measuring theory, then the effectiveness of clustering results is iterated through evolution computing during the clustering process and give the feedback to the variation of fuzzy weighting exponent m so that m receives a stable solution. Both theoretical analysis and experiments have shown that this algorithm is effective, and the fuzzy weighting exponent m corresponds with the expected results.