The application of pattern recognition for the prediction of the coal mining water irruption

Application of pattern recognition is discussed in the prediction of the water irruption from the floor of the underground coal mine. By exhaustive analysis of the hydrogeological data acquired from the coal mine, three data features are selected as the most useful out of six proposed features. Over 90% of prediction accuracy is achieved with Bayes classifier. For the reliability of the decision function, an ergodic stability test of the function is developed that shows that the reliability of the decision function is over 89%. The prediction shows that the pattern recognition method is applicable