文献类型：期刊文献

英文题名：An emotional preference ensemble clustering based on directional increment and non-linear cosine adaptive crossover and mutation

作者：Dai, Mingzhi[1,2,3];Feng, Xiang[2,3];Yu, Huiqun[2,3];Guo, Weibin[2]

机构：[1]Shanghai Jiao Tong Univ, Sch Aeronaut & Astronaut, Shanghai 200240, Peoples R China;[2]East China Univ Sci & Technol, Dept Comp Sci & Engn, Shanghai 200237, Peoples R China;[3]Shanghai Engn Res Ctr Smart Energy, Shanghai 200237, Peoples R China

年份：2025

卷号：55

期号：10

外文期刊名：APPLIED INTELLIGENCE

收录：;EI(收录号：20252518646796);WOS:【SCI-EXPANDED(收录号:WOS:001512196000021)】；

基金：This work is supported by the National Natural Science Foundation of China (No.62276097), Key Program of National Natural Science Foundation of China (No.6213-6003), National Key Research and Development Program of China (No. 2020YFB1711700), Special Fund for Information Development of Shanghai Economic and Information Commission (No.XX-XXFZ-02-20-2463) and Scientific Research Program of Shanghai Science and Technology Commission (No.21002411000).

语种：英文

外文关键词：Swarm intelligence; Ensemble learning; Non-linear position weight; Directional position increment; Non-linear cosine adaptive crossover and mutation; Data clustering

摘要：Swarm intelligence optimization has been widely utilized in many fields with the growth of scientific engineering and mathematics, and clustering is a data mining problem that requires settling theoretically. Since the emotional preference and migration behavior clustering (EPMC) model has efficiently solved the clustering tasks, it also suffers from premature convergence and the population diversity problem. Besides, the solution searching range of EPMC is too broad, and the evolution process is entirely random. Therefore, in order to enhance the optimization ability of the EPMC algorithm, we incorporate several evolution strategies such as the non-linear position weight, the directional position increment, and the novel non-linear cosine adaptive crossover and mutation method into the emotional preference and migration behavior clustering (EPMC) model. As a result, an emotional preference migration clustering model based on position increment nonlinear cosine adaptive crossover and mutation strategy (DPNG-EPMC) was proposed. The DPNG-EPMC can further prevent the algorithm from prematurely falling into local optimality similar to gradient descent, improve the diversity of the population, and enrich the family of the clustering methods. The proposed DPNG-EPMC is compared with the other nine clustering algorithms through numerous experiments, and the results of several criteria show the effectiveness of our proposed ensemble algorithm. Besides, theoretical analysis has verified the convergence of the DPNG-EPMC algorithm.