INCM: neutrosophic c-means clustering algorithm for interval-valued data

Data clustering has emerged as a prospective technique for analyzing interval-valued data and has found extensive applications across various practical domains. However, the presence of outliers and imprecise information in the real world renders fuzzy clustering cannot capture the overall information of complex data. Despite neutrosophic c-means clustering can reflect the imprecision and uncertainty and is immune to outliers, the inherent limitation lies in its capability to exclusively represent single-valued data. To tackle the above dilemma, in this paper, we propose a suitable extension of neutrosophic c-means clustering, termed as INCM, especially designed for interval-valued data. We formulate a novel objective function and provide iterative procedures for updating cluster prototype and neutrosophic partition. Finally, we conduct numerous experiments to illustrate the superiority of INCM against existing clustering algorithms on synthetic and real-world data sets.