作者: S. Buratti , S. Benedetti , M. Scampicchio , E.C. Pangerod
DOI: 10.1016/J.ACA.2004.07.062
关键词: Linear discriminant analysis 、 Electronic nose 、 Northern italy 、 Artificial intelligence 、 Electronic tongue 、 Principal component analysis 、 Multivariate statistics 、 Chemical sensor 、 Chemistry 、 Pattern recognition
摘要: Abstract An electronic tongue based on amperometric detection in a flow injection system (FIA) and commercial nose have been used to characterize classify four types of Barbera wines having different denominations origin produced northern Italy enclosed geographical areas. E-nose e-tongue data were compared elaborated together with those chemical analysis colour evaluation. All treated by multivariate processing principal component (PCA), linear discriminant (LDA) classification regression trees (CART) analysis. In the best results obtained from LDA applied data, this giving 100% correct assignation 98.1% prediction.