Classification Research on Syndromes of TCM Based on SVM

Chunming Xia, Feng Deng, Yiqin Wang, Zhaoxia Xu, Guoping Liu, Jin Xu, Helge Gewiss
{"title":"Classification Research on Syndromes of TCM Based on SVM","authors":"Chunming Xia, Feng Deng, Yiqin Wang, Zhaoxia Xu, Guoping Liu, Jin Xu, Helge Gewiss","doi":"10.1109/BMEI.2009.5305418","DOIUrl":null,"url":null,"abstract":"Syndrome is a unique TCM concept, which is an abstractive collection of symptoms and signs. Several modern algorithms have been applied to classify syndromes, but no satisfied results have been obtained because of the complexity of diagnosis procedure. Support vector machine (SVM) has been found to be very efficient to solve the classification problems, especially for binary classification with good generalization properties. In this paper, firstly patients’ clinic data of heart disease were preprocessed, then chose the optimal kernel function and used the cross-validation method to find the best parameters for SVM model, finally, the accuracy of testing different syndromes in accordance with pathology of heart disease was obtained. The results indicated that SVM was the best identifier with 81.08% accuracy on samples than the stepwise regression with 77.30% and the neural network with 73.72%. In addition, by comparing with four different kernel functions of SVM, radial basis function (RBF) was the best identifier than the others. Keywords-Syndrome; Traditional Chinese Medicine; Support Vector Machine","PeriodicalId":6389,"journal":{"name":"2009 2nd International Conference on Biomedical Engineering and Informatics","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2009-10-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2009 2nd International Conference on Biomedical Engineering and Informatics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/BMEI.2009.5305418","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 7

Abstract

Syndrome is a unique TCM concept, which is an abstractive collection of symptoms and signs. Several modern algorithms have been applied to classify syndromes, but no satisfied results have been obtained because of the complexity of diagnosis procedure. Support vector machine (SVM) has been found to be very efficient to solve the classification problems, especially for binary classification with good generalization properties. In this paper, firstly patients’ clinic data of heart disease were preprocessed, then chose the optimal kernel function and used the cross-validation method to find the best parameters for SVM model, finally, the accuracy of testing different syndromes in accordance with pathology of heart disease was obtained. The results indicated that SVM was the best identifier with 81.08% accuracy on samples than the stepwise regression with 77.30% and the neural network with 73.72%. In addition, by comparing with four different kernel functions of SVM, radial basis function (RBF) was the best identifier than the others. Keywords-Syndrome; Traditional Chinese Medicine; Support Vector Machine
基于支持向量机的中医证候分类研究
证是一个独特的中医概念,是症状和体征的抽象集合。目前已有几种现代算法应用于证候分类,但由于诊断过程复杂,结果并不理想。支持向量机(SVM)是解决分类问题的有效方法,特别是对于具有良好泛化特性的二值分类。本文首先对心脏病患者的临床数据进行预处理,然后选择最优核函数,利用交叉验证的方法寻找SVM模型的最佳参数,最后获得根据心脏病病理检测不同证候的准确率。结果表明,SVM对样本的识别准确率为81.08%,优于逐步回归的77.30%和神经网络的73.72%。此外,通过对比4种不同的支持向量机核函数,径向基函数(RBF)是最优的识别方法。Keywords-Syndrome;中医;支持向量机
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信