Multi-source Knowledge Discovery And Management Methods Of Petrochemical Enterprises

Enterprise knowledge management includes the activities of enterprise effectively identifying,fusing and utilizing the explicit and tacit knowledge with modern information technology.Efficiently applying knowledge mining methods in the process of knowledge management,can help petrochemical enterprises to conduct a better index prediction,risk analysis and enterprise decision to improve the customer loyalty and service experience degree.Thus,it will fundamentally improve the core competition of petrochemical enterprises in China.However,enterprise knowledge management system can only handle the data from single source if there are no effective multi-source information processing methods,which will inevitably limit the development space of the petrochemical enterprises.Therefore,this dissertation explores to build up a high-efficient knowledge management system,which establishes a reasonable and effective mechanism to analyze employees' character,preference and behavior mode comprehensively and thoroughly to improve enterprise level of human resource management and performance management through further studying the network behavior modeling and the methods of multi-source data combination.The main contributions of the dissertation are described as follows.Firstly,it proposes a method of information presentation and behavior modeling,based on multi-source information combination.Secondly,it proposes a method of index relevance based on multi-source information combination.Thirdly,it proposes a method of employee community analysis based on multi-source information combination.Fourthly,it proposes a method of customer index correlation analysis based on multi-source information combination,and vertical classification based on hierarchical clustering for petrochemical enterprises through the technology of big data analysis and natural language processing.Based on the above methods,the dissertation employs group experiment and contrast evaluation on the effectiveness of knowledge management in petrochemical enterprises.The results indicate that knowledge management level is greatly improved with the help of the new knowledge management methods based on the fusion analysis of multi-source and heterogeneous big data.The research results of the dissertation will provide a feasible choice for the future study of enterprise knowledge management methods.