Abstract: Coal seam gas content is not only an important parameter for the comprehensive evaluation of coalbed methane reservoirs, at the same time, accurately predicted coal seam gas content is also an important measure to prevent gas explosions.Therefore, it is very important to accurately determine the gas content of the coal seam.In response to this problem, taking the Linxing block of the eastern margin of the Ordos Basin as the research object, combined with previous research results, an integrated algorithm model based on the decision tree model was introduced.Based on the measured data in the study area, the SVM of coal seam gas content was established, neural network models was established, random forest models was established, and gradient-elevation tree models was established to predict the performance of each model.The results show that the ensemble algorithm model based on decision tree model has a better prediction effect and stronger stability, and it is more advantageous than SVM model and ANN model in the sample set with less sample size and lower dimension.