Abstract: In order to effectively predict the risk of coal and gas outburst in a coal seam disclosed section of Taozi? ya tunnel, nine key factors affecting coal and gas burst are determined in light of the comprehensive hypothesis about coal and gas burst and the Regulation of Preventing Coal and Gas Burst. Because of the complex nonlinear mapping relationship between evaluation factors and risk degree of burst, a nonlinear support vector machine (SVM) method is adopted to predict the risk of tunnel coal and gas burst. The specific parameters of all training samples are determined, a comparison of risk prediction of coal and gas burst is conducted to the selected training samples by single index method, optimal classification decision function and the MATLAB SVM Toolbox software based on the actual situation. The prediction results of the two measuring points of N7 and N8 show that there are risks of coal and gas outburst in the disclosed coal section of Taoziya tunnel, so it is necessary to take relevant measures to prevent coal outburst.