A fuzzy cluster validity index with secondary options detector based on correlation

[Submitted on 28 Aug 2023]

Download a PDF of the paper titled A correlation-based fuzzy cluster validity index with secondary options detector, by Nathakhun Wiroonsri and Onthada Preedasawakul

Download PDF

Abstract: The problem of determining the optimal number of clusters in cluster analysis is a significant concern. Existing cluster validity indexes have not adequately addressed the situation where multiple options can be chosen as the final number of clusters. In this study, we propose a new correlation-based fuzzy cluster validity index called the Wiroonsri-Preedasawakul (WP) index. This index is defined based on the correlation between the actual distance between a pair of data points and the distance between adjusted centroids with respect to that pair. We compare the performance of our index with several existing indexes on different types of datasets and find that the WP index outperforms most of them in accurately detecting the optimal number of clusters and providing accurate secondary options. Our index also remains effective when the fuzziness parameter m is set to a large value. The R package WPfuzzyCVIs, used in this work, is available at this URL: https://github.com/nwiroonsri/WPfuzzyCVIs.