摘要
提出一种基于支持向量机(SVM)的气敏传感器阵列信号处理新方法。SVM作为一种新的机器学习方法,由于其建立在结构风险最小化准则之上,从而使得支持向量分类器具有良好的推广能力。该文首先讨论了SVM的基本原理,然后将其作为识别气体种类的分类器。该方法可以得到较高的识别率,能够识别复杂的模式。
A new signal processing method for gas sensor array, based on support vector machines (SVMs),is presented. The SVM is a new machine learning method. It operates on the principle of structure risk minimization, hence a better generalization ability is guaranteed. The basic principle of the SVM is discussed at first, and then it is used as a classifier to identify the gas category. This method can classify complicated patterns and achieve higher recognition rate.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2005年第8期871-875,共5页
Chinese Journal of Scientific Instrument
基金
浙江省自然科学基金(602145)资助项目。
关键词
气敏传感器阵列
支持向量机
模式识别
Gas sensor array Support vector machines Pattern recognition