Robust-learning fuzzy c-means clustering algorithm with unknown number of clusters

作者：

Highlights：

• We construct a robust learning-based fuzzy c-means (FCM) framework, called the robust-learning FCM (RL-FCM) algorithm.

• The proposed RL-FCM can automatically find the best number of clusters, without any initialization and parameter selection with free of the fuzziness index m.

• The computational complexity of the proposed RL-FCM algorithm is analyzed.

• The experimental results and comparisons actually demonstrate these good aspects of RL-FCM where it exhibits three robust characteristics.

摘要

•We construct a robust learning-based fuzzy c-means (FCM) framework, called the robust-learning FCM (RL-FCM) algorithm.•The proposed RL-FCM can automatically find the best number of clusters, without any initialization and parameter selection with free of the fuzziness index m.•The computational complexity of the proposed RL-FCM algorithm is analyzed.•The experimental results and comparisons actually demonstrate these good aspects of RL-FCM where it exhibits three robust characteristics.

论文关键词：Fuzzy clustering,Fuzzy c-means (FCM),Robust learning-based schema,Number of clusters,Entropy penalty terms,Robust-learning FCM (RL-FCM)

论文评审过程：Received 27 August 2016, Revised 12 May 2017, Accepted 20 May 2017, Available online 22 May 2017, Version of Record 29 May 2017.

论文官网地址：https://doi.org/10.1016/j.patcog.2017.05.017