Analysis and application of clustering validity indexes
Xiao Li, Xu Zhong, Heping Peng, Hongbin Wang, Qingdan Huang · 2022 International Conference on Machine Learning and Intelligent Systems Engineering (MLISE) · 2022
Nowadays, data mining has become an important means of processing massive amounts of data, and the clustering process is unknown to the data structure, so it is difficult to clarify the optimal number of clusters in the clustering algorithm. The number of clusters is often closely related to the performance of the clustering algorithm and can be used to estimate the quality of the clustering. This paper briefly introduces the different types of clustering effectiveness indicators, follows the classification standards of the indicators, and summarizes the differences, advantages and disadvantages of the existing clustering validity index clustering validity indicators through experiments. The problems and challenges that may arise in the future of clustering effectiveness index research are discussed.