Abstract: For lack of deep analysis of hidden hazard data in current coal mine hidden hazard management, data mining algorithms which were suitable for discovering association rule of hidden hazard were introduced, and support-confidence-Kulczynski model was proposed to indicate association relationship among hidden hazard factors. Data warehouse is built after preprocessing and conversion of hidden hazard data, and mining analysis is conducted on four dimensions such as department, category, level and address of hidden hazard, so as to provide corresponding assistant decision-making according to strong association rule founded among dimensions. The actual application results show that occurrence of hidden hazard is reduced and reliable support is provided for coal mine hidden hazard management by use of the data mining algorithm.