Source discrimination of mine water based on the random forest method

Machine learning is one of the widely used techniques to pattern recognition. Use of the machine learning tools is becoming a more accessible approach for predictive model development in preventing engineering disaster. The objective of the research is to for estimation of water source using the machine learning tools. Random forest classification is a popular machine learning method for developing prediction models in many research settings. The type of mine water in the Pingdingshan coalfield is classified into surface water, Quaternary pore water, Carboniferous limestone karst water, Permian sandstone water, and Cambrian limestone karst water. Each type of water is encoded with the number 0–4. On the basis of hydrochemical data processing, a random forests model is designed and trained with the hydrochemical data. With respect to the predictive accuracy and robustness, fourfold cross-validation (CV) is adopted for the model training. The results show that the random forests model presented here provides significant guidance for the discrimination of mine water.