一种广义加权模糊聚类算法

提出了一种广义的加权模糊聚类新算法来处理具有不同特征贡献和不同数据分布的混合属性数据.分别利用样本概率密度思想和ReliefF算法为每一个样本和每一维特征分配权值,通过样本和特征的加权,将模糊c均值算法、模糊c-modes算法、模糊c-原型算法以及样本加权聚类算法统一为一个通用的框架.不同测试数据集的实验结果证明,这种广义的模糊聚类新算法对于处理不同分布以及具有不同特征贡献的大数据集是相当有效的.

A general weighted fuzzy clustering algorithm

A new general Weighted Fuzzy Clustering Algorithm is proposed to deal with the mixed data including different feature contribution and different sample distribution,in which the idea of the probability density of samples is used to assign the weights for every samples and the ReliefF algorithms is applied to give the weights for every features.By weighting the samples and their features,the fuzzy c-means,fuzzy c-modes,fuzzy c-prototype and sample-weighted clustering algorithms can be unified into a general framework.The experimental results with various test data sets illustrate the effectiveness and robustness of the proposed clustering algorithm to deal with different sample discribution and big data sets including different feature contribution.