Study On Topic Map With Mining Association Rules Based On Quantitative Concept Lattice

As a new digital knowledge representation and organization method, Topic Map has become an important research direction in the field of knowledge discovery. Topic Map can not only locate a certain knowledge, can also represent the link between knowledge points, and provide learners with the navigation of knowledge in the form of visualization. But now the theme extraction on the Topic Map has not yet find an ideal method, the traditional keyword-based matching method will lead to a high error rate in some cases.Association rules as an important research branch in the field of data mining, its main purpose of the study is mining the hidden, interesting relationship between the attributes from the large data set.Concept Lattice is hierarchical structure based on the dual relationship.It was ofen used in data analysis and mining rules.Study found Concept Lattice can reduce redundancy rules,improve the efficiency of mining rules,it is an effective method in mining association rules.In this paper, we adopt Topic Map technology to organize effectively teaching resources on the network environment, the Topic Map technology use association rules mining based on Quantitative Concept Lattice. For the shortcomings of the common theme extraction method of the Topic Map, this paper make full use of the advantages of mining association rule based on Quantitative Concept Lattice, and the method was used in the theme extraction. Compared to the traditional keyword-based matching method, the method in this paper has improved the similarity accuracy between the keywords. The main research work is reflected in the following aspects:(1) Design a new model TMTRM on the basis of the common model of Topic Map, which combine with the features of network resources, and implement the model through the Topic Map technology of mining association rules based on Quantitative Concept Lattice. The experiments showed that the model can effectively organize and manage teaching resources on the Web,and realized personalized retieval and knowledge recommendation.(2) Propose a method that is mining association rules based on Quantitative Concept Lattice, which will be used in extracting themes, and the experiment have proved that it can improve significantly the correct rate of the extracted keywords compared to traditional methods.(3) Build a personalized Topic Map with mining association rules based on Quantitative Concept Lattice, then test the performance of personalized retrieval based on the Topic Map, and evaluate the rationality of the knowledge recommendation algorithm of mining association rules based on Quantitative Concept Lattice. Experiments show that personalized search has been greatly improved in precision.