中文说明：模糊聚类是一种重要数据分析和建模的无监督方法.在FCM算法中,考虑到样本矢量中各维特征对模式分类的不同影响,本文提出一种优化特征加权的模糊聚类算法,该算法利用主成分分析法提取主要特征向量并根据其对方差的贡献率不同赋予相应权重进行聚类分析.
English Description：
Fuzzy clustering is an important unsupervised method for data analysis and modeling. In FCM algorithm, considering the different influence of each dimension feature in sample vector on pattern classification, this paper proposes a fuzzy clustering algorithm with optimized feature weighting. The algorithm extracts main feature vectors by principal component analysis and gives corresponding weights according to their contribution rate to variance for clustering analysis