Hierarchical Normal Wiggly Hesitant Fuzzy K-means Clustering Algorithm

Normal wiggly hesitant fuzzy sets (NWHFSs) as a significant tool in decision-making. NWHFSs can retain the original information and dig the uncertain information based on hesitant fuzzy information. However, there are few studies in clustering of NWHFSs. In this paper, we propose a method for NWHFSs based on hierarchical clustering combine with K-means algorithm. Firstly, we get the hesitant fuzzy decision matrix by decision makers (DMs). Meanwhile, we mine the potential information and obtain the normal wiggly hesitant fuzzy decision matrix. Then, we calculate the distance between patterns and repeat it until all patterns are clustered into one. Furthermore, we select the results of hierarchical clustering as initial cluster and calculate centroids. Furthermore, we redistribute the patterns and repeat these steps until the centroids is unchanged. Finally, we use an example about the evaluation in flow environment of data element to illustrate the particularly and validity of the proposed method.