一种改进的FLS-SVM分类辨识模型及其应用

来源期刊:中南大学学报(自然科学版)2017年第8期

论文作者:左红艳 王涛生

文章页码:2097 - 2105

关键词:混沌免疫算法;模糊最小二乘支持向量机;分类辨识

Key words:chaos immune algorithm; fuzzy support vector machines; classification identification

摘    要:采用三角形函数隶属度法确定模糊最小二乘支持向量机(fuzzy least squares support vector machine, FLS-SVM)输入参数隶属度,采用自适应变尺度混沌免疫算法优化FLS-SVM的参数,从而构建改进模糊最小二乘支持向量机(improved fuzzy least squares support vector machines, IFLS-SVM)分类辨识模型, 用Ripley数据集、MONK数据集和PIMA数据集进行仿真实验,并用于地下金属矿山采场信号分类辨识与中国国际贸易安全分类辨识。研究结果表明:与LS-SVM分类辨识模型和FLS-SVM分类辨识模型相比,IFLS-SVM分类辨识模型能有效提高带噪声点和异常点数据集的分类精度,且分类辨识精度相对误差较小。

Abstract: A classification and identification model was developed based on improved fuzzy least squares support vector machines(FLS-SVM),in which the fuzzy membership function was set by using triangle function method and its parameters were optimized by an adaptive mutative scale chaos immune algorithm, and an improved fuzzy least squares support vector machines(IFLS-SVM) was constructed. The simulation experiments were conducted on three benchmarking datasets such as Ripley datasets, MONK datasets and PIMA datasets for testing the generalization performance of the classification and identification model, signals from underground metal mines stope wall rock and international trade data in China were diagnosed by the IFLS-SVM classification and identification model. The results show that compared with LS-SVM classification identification model and FLS-SVM classification identification model, the IFLS-SVM classification identification model is valid for improving the analysis accuracy of the data with noises or outliers and IFLS-SVM classification identification model has small relative error.

相关论文

  • 暂无!

相关知识点

  • 暂无!

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号