Abstract: In order to improve the accuracy to predict gas content of seam, based on gas content in vertical depth with interval of 50 m in a mining area, a linear do uble exponential smooth method was applied to establish prediction model to predict gas content and in combination with polygonal line graph and variance analysis, a comparison was conducted on the predicted results and the measured data. The results showed that the linear double exponential smooth method model with a smooth constant a of 0. 8 could have high prediction accuracy. The application of the linear double exponential smooth method to predict the gas content could meet a general t endency of the gas content increased in linear law with the depth increased and also could accurately predict the seam gas content based on the comparison on the up dated measured gas content. Based on the measured values involved in the weight law of the model, the model could approach the measured data of a new round.