中文摘要：

基于局部线性嵌入（Locally Linear Embedding ，LLE）算法和极限学习机（Extreme Learning Machine ， ELM）神经网络建立矿井瓦斯涌出量预测模型，该预测模型运用 LLE 算法对矿井瓦斯涌出量影响因素样本进行数据挖掘，得到降维后的有效因子，再将这些有效因子作为 ELM 神经网络的输入层进行训练和预测。利用某矿井的实测数据进行实例分析，结果表明该预测模型预测速度快，精度高，能够用于矿井瓦斯涌出量预测。

英文摘要：

Based on locally linear embedding （LLE） algorithm and extreme learning machine （ELM） neural network ,a mine gas emission prediction model is established .The prediction model uses LLE to obtain effective factors by mining effective fac-tors from sample data and then as the input layer in ELM neural network ,the effective factors are trained and predicted .An analysis of the measured data of a certain coal mine is conducted and the results show that the prediction model has the fast speed of prediction ,high precision and can be used for prediction of mine gas emission .