Abstract: In order to systematically analyze the risk factors of gas explosion and promote the safe and efficient production of coal mines, the research first builds a coal mine gas explosion risk factor index system based on four dimensions of human factors, environmental conditions, production facilities and management organizations. Secondly, the particle swarm optimization algorithm is used to optimize the BP neural network to obtain the direct correlation matrix between risk factors, and then the DEMATEL method is used to calculate the influence degree, the influenced degree, centrality and cause degree are analyzed to obtain strong driving risk factors and driving risk factors, and the entropy weight method is used to calculate the comprehensive importance of each risk factor, and finally get the key risk factors of coal mine gas explosion according to the sorting results. The results show that: safety education training and implementation, safety management organization, safety management culture, employee safety awareness, safety technology and investment are the keys to gas explosion disaster risk prevention.