Mining hidden danger data association rules of coal mining face based on genetic algorithm

This study delved into the classification and attributes of potential hazards within coal mining operations, utilizing a genetic algorithm to develop an association rule mining model. By integrating text mining and topic mining algorithms, it uncovered the intrinsic relationships and association rules among identified hazards, leading to the creation of an association rule database. Utilizing safety hazard inspection records from a mining company in Shandong Province as a data source, the model underwent rigorous validation. Furthermore, a comparative analysis of the performance between the enhanced genetic algorithm and both the original genetic and Apriori algorithms was conducted. The findings demonstrate that the refined genetic algorithm is markedly efficient in uncovering the association rules within hidden danger data, thereby significantly enriching safety managers' understanding of the underlying patterns among these data points. This enhanced insight serves as a solid foundation for the detection and remediation of safety hazards in coal mines, ultimately contributing to the advancement of safety management practices in coal mining operations. © 2024 Editorial Office of Journal of Mining and Strata Control Engineering. All rights reserved.