A novel fuzzy C-means algorithm to generate diverse and desirable cluster solutions used by genetic-based clustering ensemble algorithms

摘要: One of the most significant discussions in field machine learning today is on clustering ensemble. The ensemble combines multiple partitions generated by different algorithms into a single solution. Genetic are known for their high ability to solve optimization problems, especially problem To date, despite major contributions find consensus cluster with application genetic algorithms, there has been little discussion population initialization through generative mechanisms genetic-based as well production favorable fitness values first phase ensembles. In this paper, threshold fuzzy C-means algorithm, named TFCM, proposed diversity clustering, one common problems Moreover, TFCM able increase partitions, such that it improves performance algorithms. average evaluated three objective functions and compared against other simple SGCE, proposed, which used initial SGCE. SGCE based populations used. experimental results eleven real world datasets demonstrate enhanced using TFCM.