Research On Matching Supply And Demand Of Enterprise Tacit Knowledge Exposure Cases In Knowledge Service Context

In the new economy,tacit knowledge resources have gradually evolved into the core competitive advantages of enterprises.However,the rapid enrichment of enterprise knowledge resources has given rise to the problem of "knowledge disorientation",resulting in the low utilization rate of tacit knowledge resources.Guided by users’ knowledge demand,tacit knowledge supply and demand matching retrieves knowledge matching with users’ demand from rich knowledge resources to ensure the provision of fast and accurate knowledge services to users.However,in practical application,due to the limitation of the huge knowledge volume and the algorithm of full library traversal,the time cost of knowledge matching continues to increase and the matching efficiency is low;moreover,the traditional single measurement standard(view similarity)can no longer meet the users’ needs.In view of this,this thesis takes tacit knowledge episodic cases as the research object and implements re-organization of the tacit knowledge episodic case base by improving the FCM algorithm to improve the algorithm operation efficiency;at the same time,the traditional view similarity method is improved by integrating the similarity and relevance between tacit knowledge episodic cases to improve the knowledge service benefits.Firstly,a case-based knowledge representation method is used to externalize the tacit knowledge,and a fuzzy meta-analysis method is applied to characterize the user’s knowledge demand;then an improved FCM algorithm is used to classify clusters,and the traversal space is compressed horizontally to improve the efficiency of the algorithm;for the problem that the initial number of clusters determined by the FCM algorithm has a certain randomness,a DBSCAN algorithm integrating density and distance metric mechanism is used to initialize the cluster centers,and an improved cosine similarity optimization objective function is introduced to determine the optimal number of clusters to improve the performance of the clustering algorithm and thus enhance the efficiency of knowledge supply and demand matching.And then,the Zadeh-PFS-based method is used to calculate the matching degree between user requirements and the cases of tacit knowledge extrapolation.Considering the influence of inter-knowledge correlation on the matching results,the view similarity and correlation are integrated to construct a knowledge matching degree model;meanwhile,the coefficient of variation method-CRITIC assignment method is selected for weight calculation,and the game theory idea is borrowed to enhance the effectiveness of combined assignment;on this basis,the improved Zadeh operator similarity and PFS correlation coefficient are adopted to measure the matching degree between user knowledge needs and existing.On the basis of this,the thesis adopts the improved Zadeh operator similarity and PFS correlation coefficient to measure the matching degree between users’ knowledge requirements and existing knowledge,compares them with the matching threshold,and selects the knowledge with the best matching degree with users’ requirements for submission to users,and also assists in recommending other knowledge above the matching threshold to support users to implement case adjustment or multi-case adaptation.Finally,a comparative validation is conducted based on the Wine quality dataset in the UCI database to test the effectiveness of the proposed knowledge supply and demand matching method.The experimental results are analyzed and compared with those of the traditional matching model,the traditional Zadeh operator without considering the association property among knowledge,and the matching model with improved Zadeh operator,respectively,and the experimental results show that the proposed algorithm has certain comparative advantages in terms of accuracy and efficiency.