Kernel C-Means Clustering Algorithms for Hesitant Fuzzy Information in Decision Making

When facing clustering problems for hesitant fuzzy information, we normally solve them on sample space by using a certain hesitant fuzzy clustering algorithm, which is usually time-consuming or generates inaccurate clustering results. To overcome the issue, we propose a novel hesitant fuzzy clustering algorithm called hesitant fuzzy kernel C-means clustering (HFKCM) by means of kernel functions, which maps the data from the sample space to a high-dimensional feature space. As a result, the differences between different samples are expanded and thus make the clustering results much more accurate. By conducting simulation experiments on distributions of facilities and the twenty-first Century Maritime Silk Road, the results reveal the feasibility and availability of the proposed HFKCM algorithm.