Application of Least Squares Support Vector Machines for Discrimination of Red Wine Using Visible and Near Infrared Spectroscopy

摘要：

Visible and near infrared(Vis/NIR)transmittancespectroscopy and chemometrics methods were utilizedto discriminate red wine.The samples of five varietiesof red wine were separated into calibration set andvalidation set randomly.The principal components(PCs)could be obtained from original spectrum byusing Partial least squares(PLS),The PCs(selectedby PLS)of each sample in calibration set was used asthe inputs to train the Least squares support vectormachines(LS-SVM)model,then the optimal modelwas used to predict the varieties of samples invalidation set based on their PCs,and 94％recognition ratio was achieved with the thresholdpredictive error ±0.1,while 100％ recognition rationwith the threshold predictive error ±0.2.Root meansquare error of prediction(RMSEP)and determinationcoeefficient(r2)were O.0531 and O.9986 respectively.Itis indicated that Vis/NIR transmittance spectroscopycombined with PLS and LS-SVM is an efficientmeasurement to discriminate types of red wine.

作者: Fei Liu Li Wang Yong He

作者单位: College of Biosystems Engineering and Food Science,Zhejiang University,268 Kaixuan Road,Hangzhou 310029,China

会议类型: 国际会议

会议名称: 2008 3rd International Conference on Intelligent System and Knowledge Engineering(第三届智能系统与知识工程国际会议)(ISKE 2008)

会议地点: 厦门

会议语种:英文

页码: 1002-1006

在线出版日期: 2008-11-17（万方平台首次上网日期，不代表论文的发表时间）