基于2型模糊集的粗糙模糊C-means算法

Type-2 Fuzzy Set based Rough Fuzzy C-means Clustering Algorithm

聚类算法在图像处理、模式识别等领域有广泛应用,粗糙模糊C-means算法是近年来研究较多的聚类算法。在面对聚类结构不同的样本时,传统的粗糙模糊C-means算法存在聚类簇心偏向性和隶属度选取的问题,使聚类结果不理想。提出一种基于②型模糊集的粗糙模糊C-means算法,算法采用②型模糊集理论,计算样本的次隶属度,从而描述样本的深层信息,根据样本最大隶属度和次大隶属度之间的差别,将样本划分到类簇的上下近似集中,根据上下近似集的权重,迭代并重新计算簇心,直到达到设定阈值或者满足算法终止条件。将改进的粗糙模糊C-means算法在人工数据集和UCI数据集上进行实验对比,结果表明改进的粗糙模糊C-means算法具有良好的性能。

Clustering algorithm is widely used in image processing,pattern recognition and other fields.The RFCM algo-rithm is a clustering algorithm that has been studied more in recent years、When the clustering structure of sample is dif-ferent,the traditional RFCM algorithm has the problem of cluster center bias and membership selection,which makes the clustering result worse.This paper proposes a RFCM C-means algorithm based on type-2 fury set.The refined RFCM algorithm uses the type-2 fuzy set theory to describe the deep information of the sample by calculating the sub-degree of membership of the sample.The sample is divided into the upper and lower approximation sets of the cluster based on the difference between the maximum membership degree and the second-largest membership degree,and according to the weights of the upper and lower approximation sets,the cluster center is iterated and recalculated until the set threshold is reached or the algorithm termination condition is met.,The performances of improved RFCM C-means algorithm experi-mented on the artificial datasets and the UCI datasets are compared,the results show that the improved RFCM algorithm has good performance.