K-medoids method based on divergence for uncertain data clustering
Jin Zhou, Yuqi Pan, C. L. Philip Chen, Dong Wang, Shiyuan Han · 2016
Uncertain data clustering is an essential task in the research of data mining. Lots of traditional clustering methods are extended with new similarity measurements to tackle this issue. Different from certain data clustering, uncertain data clustering focus more on the evaluation of distribution similarity between uncertain data objects. In this paper, based on the KL-divergence and the JS-divergence, we propose a novel K-medoids method for clustering uncertain data, named UK-medoids. Good performance of the proposed algorithm is shown in experiments on synthetic datasets.