Research And Application Of Proactive Knowledge Service Technology Based On Task-driven

With the development of knowledge economy, the innovation based on knowledge resources has been becoming the core competitiveness in enterprise. Enterprises have accumulated vast amounts of heterogeneous knowledge resources in the process of information technology. However, with the expansion of knowledge, it is difficult for users to get the necessary knowledge accurately and effectively, which has led to many problems such as "knowledge-overload" and "knowledge-trek” and so on. To solve these current difficulties of the modeling of knowledge requirement in the process of proactive knowledge service, and improve the accuracy of knowledge recommendation, the technology of task-based model of knowledge requirement automatic modeling and the task-based algorithm of collaborative filtering recommendation are deeply studied. These main researches of the thesis are as follows:(1)The current research status of knowledge service are analyzed,and the framework of a proactive knowledge service system based on task-driven is designed, then the system platform’s knowledge service process is given.(2)An automatic modeling method of task-based model of knowledge requirement is suggested and automatic modeling process and method of the model of knowledge requirement is designed. Firstly, the method acquire the best knowledge node by the ontology matching process based on artificial immune, then rule-based knowledge merging process is used to establish the mapping relationship between the task knowledge and the best knowledge node, and insert it into the knowledge requirement model to complete the process of automatic modeling. Finally, experimental results verify the effectiveness of the modeling method.(3)A task-based technology of collaborative filtering recommendation is studied and task-based process and method of collaborative filtering recommendation is designed. First of all, the method filters related knowledge in the model of knowledge requirement for task by the process of knowledge filtering, and then the generating process of knowledge recommending list are suggested to perform the progress of recommendation and generate the prioritized recommending list to the target user. Experimental results show that the technology has better recommending accuracy and quality.(4) The project of task-driven based proactive design knowledge service system is applied into the knowledge management system in the system of rapid design system for railway freight trunks, and the project background is introduced, then main functionalities of project are designed and implemented, and finally system application results are given.