Monitoring of methane gas is one of the important factors affecting the coal mine safety. The online real-time monitoring of the methane gas is used for the mine safety protection. To improve the accuracy of model analysis, in the present paper, the author uses the technology of infrared spectroscopy to study the gas infrared quantitative analysis algorithm. By data mining technology application in multi-component infrared spectroscopy quantitative analysis algorithm, it was found that cluster analysis partial least squares algorithm is obviously superior to simply using partial least squares algorithm in terms of accuracy. In addition, to reduce the influence of the error on the accuracy of model individual calibration samples, the clustering analysis was used for the data preprocessing, and such denoising method was found to improve the analysis accuracy.