Study On Tacit Knowledge Explicit Case Attribute Reduction For Dynamic Data Sets

In the era of knowledge economy,knowledge has become the main source of social value and wealth.The importance of data as a carrier of knowledge in production practice is unshakable.At present,the data collected by people usually exhibits characteristics such as massive,dynamic,low-definition and rich in type,which contains precious tacit knowledge.In the application practice of knowledge management,people face not only the challenges of data collection,knowledge discovery,but also the problem of knowledge refinement.In view of this,based on the explicit manifestation of the tacit knowledge implementation case,this paper further studies the problem of attribute reduction of tacit knowledge explicit cases.In this way,the accuracy of the tacit knowledge explicit case is improved,and the application benefit of the tacit knowledge is ensured.First of all,this paper summarizes the research status at home and abroad,and analyzes the problems existing in the current research;further,the technical route and innovation points of this paper are proposed.On this basis,this paper introduces the related theoretical foundations of tacit knowledge and knowledge management,case reasoning and attribute reduction,incremental learning theory and rough sets.Thereafter,the problem of inefficient processing is caused by the conventional rough set and its existing improved method for batch processing data.This paper proposes an attribute reduction algorithm combining rough set and incremental learning,which greatly improves the efficiency of attribute reduction.Among them,based on the traditional fuzzy rough set,this paper uses the concept of relative identification relationship to represent the mechanism of attribute reduction of tacit knowledge explicit cases,based on which the attribute update criterion is defined;further,based on the incremental learning theory,design Two incremental attribute reduction algorithms.1 For dynamic case set,incremental attribute reduction algorithm I: With the addition of the case subset,first update the relative identification relationship between the attribute and the attribute set,and then execute the attribute update criteria.When all the case subsets are added,you can get Complete case set reduction.2Incremental attribute reduction algorithm II: When there is a case subset added,only the relative identification relationship between the condition attribute and the attribute set is incrementally updated.When all the case subsets are added,the relative identification relationship of the entire case set can be obtained.Then,the attribute update criteria are executed,and the reduction of the entire case set is finally obtained.Finally,the experimental analysis results show that the incremental attribute reduction algorithm proposed in this paper is more advanced than the traditional algorithm in terms of running time,number of selected attributes and classification accuracy.