An Improved KFCM Algorithm Based on Artificial Bee Colony

Kernel fuzzy C-means (KFCM) clustering Algorithm is one of the most widely used methods in data mining, but this algorithm still exists some defects, such as the local optima and sensitivity to initialization and noise data. Artificial bee colony (ABC) is a very simple, robust, stochastic global optimization tool which is used in many optimization problems. In this paper, an improved KFCM algorithm based on ABC (ABC-KFCM) is proposed. It can integrate advantages of KFCM and ABC algorithm. According to the test. compared with the FCM and KFCM clustering algorithm, the proposed algorithm improves the optimization ability of the algorithm, the number of iterations is fewer, and the convergence speed is faster. In addition, there is also a large improved in the clustering result.