A c-means clustering approach based on cloud model

A cloud model based c-means clustering approach (CMCM) is presented in this paper. Each cluster in CMCM is modeled by a cloud model, which characterizes the fuzziness and randomness of cluster and makes the clustering process more applicable than FCM. In the meantime, the new approach can avoid a trivial solution of the object function in FCM by removing the normalization from FCM. Experiments for the comparison between CMCM and other clustering algorithms including FCM show the effectiveness and efficiency of the proposed method.