Support Vector Machine Applied in the Classification of Fanjing Mountain’s Tea

Abstract

The problems of classification and regression are no longer state-of-the-art ones, however, with the popularization and extensive application of computers, especially, the rapid development of machine learning and data mining gives them fresh meaning. Nowadays, these problems become major problems that have triggered a broad range of research activities again. Due to its advantages, such as superior generalization performance, global convergence, sample dimension insensitivity, and so on, support vector machine (SVM) has made great progress in theory and application. This article compares SVM with BP neutral network to classify teas of Fanjing Mountains in view of accuracy and mean square error (MSE), and the simulation results show that the accurate rate of SVM is higher than that of BP neutral network, and the MSE of the former is lower than that of the latter.