Aiming at the shortcomings of the sparrow search algorithm, such as falling into local optimum easily and slow convergence speed, an improved sparrow search algorithm based on multi-strategy fusion is proposed. The elite chaotic reverse learning strategy is used to generate the initial population, which enhances the quality of the initial individuals and the diversity of the population, and realizes the exploration of more high-quality search areas to improve the local extremum escape ability and convergence performance of the algorithm. Combined with the random following strategy of the chicken swarm algorithm, the position updating process of the followers in the sparrow search algorithm is optimized, and the local development performance and global search ability of the algorithm are balanced. The Cauchy-Gauss mutation strategy is used to improve the ability of maintaining population diversity and resisting stagnation. Ten benchmark test functions with different characteristics are optimizated. The test results and Wilcoxon's signed rank test results both show that the improved algorithm has better optimization accuracy, convergence performance and stability. Finally, the improved algorithm is used to optimize the parameters of the least square support vector machine and is applied to the identification of coal and gas outburst risk. The effectiveness of the improved strategy and the superiority of the improved algorithm are further verified by comparetive experiments.