Abstract: Coal and gas outburst is one of the most destructive coal-rock dynamic disasters, posing a serious threat to the safe production of coal mines.Accurate prediction of gas outbursts can effectively prevent their occurrence.Due to the unclear mechanism of gas outbursts, how to accurately predict them has become a pressing issue to be resolved.This paper proposes a multi-indicator-based XGBoost algorithm for predicting coal and gas outbursts.The results show that XGBoost algorithm can not only accurately predict coal and gas outbursts but also display the weight proportions of each evaluation indicator in the prediction process.