Design of RBF Neural Network Based on Improved Canopy-K-means algorithm - 文章详情页

作者：

Lijie, Jia (Lijie, Jia.) | Wenjing, Li (Wenjing, Li.) | Junfei, Qiao (Junfei, Qiao.) (学者：乔俊飞)

收录：

摘要：

To　solve　the　problem　that　RBF　neural　network　parameters　are　difficult　to　determine,　an　improved　Canopy-K-means　algorithm　is　proposed　to　optimize　the　RBF　neural　network.　By　using　the　density-based　Canopy　algorithm　to　roughly　cluster　the　data,　the　cluster　centers　of　the　K-means　algorithm　are　initialized,　meanwhile　the　number　of　cluster　centers　is　obtained,　and　the　Canopy-K-means　algorithm　based　on　the　density　of　samples　(CKD)　is　used　to　optimize　the　RBF　neural　network.　Three　experiments,　including　nonlinear　function　approximation,　classification　of　UCI　website　datasets,　the　effluent　ammonia　nitrogen　(NH4-N)　prediction　in　wastewater　treatment　process　were　used　to　verify　the　effectiveness　of　the　algorithm.　The　results　showed　that　CKDRBF　network　had　high　classification　accuracy,　strong　approximation　ability.　©　2020　ACM.

关键词：

Ammonia Approximation algorithms Big data Classification (of information) Cluster computing Effluents Effluent treatment K-means clustering Radial basis function networks Wastewater treatment

作者机构：

[ 1 ] [Lijie, Jia]Beijing Key Laboratory of Computational Intelligence and Intelligent System, College of Electronic and Control Engineering, Beijing University of Technology, Beijing, China
[ 2 ] [Wenjing, Li]Beijing Key Laboratory of Computational Intelligence and Intelligent System, College of Electronic and Control Engineering, Beijing University of Technology, Beijing, China
[ 3 ] [Junfei, Qiao]Beijing Key Laboratory of Computational Intelligence and Intelligent System, College of Electronic and Control Engineering, Beijing University of Technology, Beijing, China

通讯作者信息：

电子邮件地址：

查看成果更多字段

成果类型
所属机构

所有年份指定年份从至