文献类型：期刊文献

英文题名：A Novel Spectral Ensemble Clustering Algorithm Based onSocial Group Migratory Behavior andEmotional Preference

作者：Dai, Mingzhi[1,2]; Feng, Xiang[1,2]; Yu, Huiqun[1,2]; Guo, Weibin[1]

机构：[1] Department of Computer Science and Engineering, East China University of Science and Technology, Shanghai, 200237, China; [2] Shanghai Engineering Research Center of Smart Energy, Shanghai, 200237, China

年份：2022

卷号：13370 LNAI

起止页码：316

外文期刊名：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

收录：EI(收录号：20223112462873)

基金：Acknowledgements. This work was supported in part by the National Natural Science Foundation of China under Grant NOs. 61772200, Shanghai Pujiang Talent Program (17PJ1401900), the Information Development Special Funds of Shanghai Economic and Information Commission under Grant NO. XX-XXFZ-02-20-2463, and the Key Program of National Natural Science Foundation of China (62136003).

语种：英文

外文关键词：Cluster analysis - Data mining - Evolutionary algorithms - Intelligent systems - Laplace transforms - Learning systems - Monte Carlo methods - Optimization - Statistics

摘要：Clustering is an unsupervised machine learning technique for data mining to find objects with similar characteristics in a group. However, due to the lack of relevant prior information on the data, numerous single models or methods cannot identify the shape and size of the clusters. Therefore, an ensemble of multiple weak models is required to further mine the implicit information of the data and improve the clustering accuracy. LSMC-EPMC is an evolutionary clustering algorithm that consists of three parts, the emotional preference and migration behavior clustering (EPMC) model, the Laplacian spectral clustering model, and the Monte Carlo statistical data simulation model. This paper mainly integrates the spectral clustering model and the Monte Carlo statistical data simulation method into the EPMC algorithm by mapping the individual in EPMC and the optimized center point in the other two methods. Through numerous experiments, LSMC-EPMC shows a significantly increased performance to EPMC and is highly competitive with the other seven clustering algorithms on several standard datasets. ? 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.