基于AdaBoost组合学习方法的岩爆分类预测研究

针对岩爆等级划分问题，考虑了岩爆灾害发生的多种主要影响因素，采用新的数据挖掘方法AdaBoost（即Adaptive Boosting）的组合学习方法，结合流行的人工神经网络BP算法，构建了集成神经网络AdaBoost-ANN（简称AB-ANN）的岩爆等级多分类预测模型。该模型克服了单一弱分类器的不稳定性，提高了分类器精度，实验结果表明，预测的结果与实际值比较吻合，证明了该方法的可行性。

Classification and prediction of rockburst using AdaBoost combination learning method

Rockburst is one of the most important geologic hazards. To solve the problem of classification and prediction of rockburst, the main factors of rockburst occurred and the effective classification methods should be considered., A new method is proposed based on combination of artificial neural networks (ANN) classifiers as weak classifiers by using AdaBoost algorithm in data mining. The AdaBoost-ANN models are established. Overcoming the instability of single classifier, the models can give more accurate and stable classification for the novel conditions. The results show that this method is reliable, constrictive and promising.