Improvement of clustering by modification of degrees of fuzziness

Due to fast growth in technology, conventional classification methods are limited in their ability to support medical diagnostics without introducing considerable ambiguities. Since the conditions are vague in medicine the fuzzy methods may be more helpful rather than crisp ones. Classification depends on number of attributes, number of clusters to be classified and index of the clustering algorithm. Because it is not possible to reduce number of attributes and clusters, therefore changing the index value is a choice to improve performance. The objective of this paper is to analyze the improvement in terms of number of iterations taken, algorithm performance and percentage of correctness of Thyroid samples and wine samples classification by modifying the index of the algorithm.