A Novel Fuzzy Clustering Algorithm Based on K-Means Algorithm

Abstract

Conventional K-means algorithm cannot obtain elevated clustering specific rate, and without difficulty be exaggerated by clustering center random initialized and remote points, however the algorithm is straightforward with low time difficulty, and can process the large data set rapidly. This paper suggests an enhanced K-means algorithm named PKM. PKM is based on similarity degree among data points made by cumulated K-means, and get the final clustering partition via fuzzy precise rate of clustering higher, and reduce the effects made by isolated points and random clustering center, at the same time, can be familiar with isolated points better. Experiments with analog information and genuine data make obvious its benefit.