Classification of Chinese Herbal medicines based on SVM

Quantitative analysis of Chinese Herbal Medicines (CHMs) is important for quality consistency evaluation. However, due to various harvesting factors and plant processing it is a challenging task. This paper presents experimental-based evaluation of five kinds of herbs using an electronic tongue and support vector machine (SVM). It was found that the performance of SVM in classifying of these herbs was superior to other selected methods. The SVM was implemented by selecting a radial basis kernel function for data mapping and conducting a parameter optimization with the K-fold cross validation. The average accuracy of the final classification can reach to 96.67% and therefore, it is feasible to employ electronic tongue and SVM method for species identification of CHMs.