基于上下文约束的噪声模糊聚类算法

摘要

针对带噪声数据的聚类问题,提出一种基于上下文约束的噪声模糊聚类算法.该算法基于标准的模糊C-均值聚类理论,在修改模糊聚类目标函数的同时,结合问题的实际背景引入上下文模糊集,修改模糊划分空间的约束条件,以减少噪声对聚类结果的影响.实验结果表明:该算法能够有效地避免噪声对聚类的影响,具有很强的鲁棒性.%A noise fuzzy clustering algorithm by context constraints is proposed for clustering problems with noise data. Based on the standard fuzzy C-means clustering theory, this paper modify the objective function for fuzzy clustering and the fuzzy partition space by introduction of the context fuzzy sets with background of practical problems, in order to reduce the impact of noise on the clustering results. Experimental results show that the algorithm can effectively avoid the impact of noise on the clustering, and with strong robustness.