Fuzzy c-means clustering based on weights and gene expression programming

Data clustering is a necessary process in many scientific disciplines, and fuzzy c-means (FCM) is one of the most popular clustering algorithms. Recently, distributing weight values and avoiding local minimization are the possible ways to improve the results of FCM. In this paper, fuzzy C-means clustering based on weights and gene expression programming (WGFCM) is proposed to improve the performance of FCM. A new weight vectors calculation based on entropy is introduced to measure distance accurately. Moreover, gene expression programming (GEP) is employed to determine the appropriate cluster centers. Experiments are conducted with ten UCI data sets to compare the proposed method with FCM. In addition, WGFCM is compared with other FCM based methods and different clustering approaches published for a fair assessment. The results show that the proposed method is far superior to FCM-based methods in terms of purity, Rand Index, accuracy rate, objective function value and iterative cost. Moreover, it has an advantage over other clustering approaches in terms of the accuracy.