Medicine composition analysis based on PCA and SVM

Abstract

Medicine analysis becomes more and more important in production and life, especially, the composition analysis of medicines. Available data are often characterized by the data with small amount and high dimensionality. Support vector machine (SVM) is an ideal algorithm for dealing with this kind of data. This paper presents a combined method of principal component analysis (PCA) and least square support vector machine (LS-SVM) to deal with the work of medicine composition analysis. The proposed method is applied to practical problems. Experiments demonstrate the predominance of the proposed method on both running time and prediction precision. © Springer-Verlag Berlin Heidelberg 2005.