Research on a New Automatic Generation Algorithm of Concept Map Based on Text Clustering and Association Rules Mining

Abstract

As an important teaching tool of visualization, the concept map has become a hot spot in the field of smart education. The traditional concept map generation algorithm is hard to guarantee the construction process and quality because of the huge amount of work and the great influence of the expert experience. A TC-ARM algorithm for automatic generation of hybrid concept map based on text clustering and association rules mining is proposed. This algorithm takes full account of the attributes of the relationship between concepts, uses text clustering technology to replace the relationship between artificial mining concepts and test questions, combines association rules mining methods to generate the concept maps, and introduces consistency of answer record parameter to improve the efficiency of concept map generation. The experimental results show that the TC-ARM algorithm can automatically and rapidly construct the concept map, which not only reduces the impact of outside experts, but also dynamically adjusts the concept map based on the basic data. The concept map generated by the TC-ARM algorithm expresses the relationship between the concepts and the degree of closeness through the relationship pairs and relationship strength, and can clearly show the structural relationship between concepts, provide instructional optimization guidance for knowledge visualization.