Research on A Coal Mine Safety Knowledge Model Based on Spatial Knowledge Map Service

Research on conceptual models for coal mine safety management rely on data mining. Text mining of big data is used to extract the laws and rules that are implied in safety models, making these laws and rules more transparent by stating them explicitly and providing visualization. Currently, the ability to process information on coal mine safety management is insufficient for resolving the problems of inefficiency and a lack of knowledge. In this paper, based on a knowledge map service, the analysis of coal mine safety maps for management purposes and the acquisition of situational factors in a multidimensional space for innovation is divided into the current situation and the expectations hidden in the data. Multidimensional situations and the role of each component’s situation require specific analysis and discussion of the issues of human-computer interactions and situations. Five levels of users' spatial knowledge demand are proposed and analyzed by combining the coal mine safety demand, multidimensional situational factors and spatial knowledge map service system. The service process model of user space knowledge acquisition and the organizational structure of the service system are constructed. Then, the coal mine safety text is classified, and a self-adaptive weight deep learning classification algorithm is used to classify the coal mine safety knowledge data to verify the difference in the classification effect of two classification algorithms. Finally, based on the correct classification of coal mine safety knowledge data, the mining of hazard association rules is performed. From the perspectives of knowledge workers, knowledge bases, and the combination of knowledge workers and knowledge bases, the mode of knowledge analysis of coal mine safety maps and its specific applications in intelligent searches, knowledge sharing and knowledge optimization are discussed.