The System Of Topic Selection In Junior Middle School Based On Knowledge Map And Its Application

In the traditional education model, it is difficult for teachers to arrange exercises based on student’s individual knowledge level. After the new curricular reform, new exercise system combining the advanced computer technologies need to be provided for student, which can recommend exercises suiting for individual knowledge level. Therefore, this paper use knowledge graph technology, and combined with the characteristics of high school students and educational theory, designs and implements a junior high school English question selection system to improve student’s learning motivation and learning efficiency, and to help teachers to know the student knowledge level, which can help improve teaching behaviors.This paper constructs the question knowledge graph through natural language processing technology and information extraction methods. First, this paper obtains an amount of question data through analyzing electronic documents and uses web crawler to get authorized or free data resources and uses NLP tools to segment, POS tagging and named entity recognition and et al. Then, this paper extracts the knowledge points based on SVM model, which involves building multi-class feature vectors and constructing a hierarchical classification for question data. After extracting knowledge points, use similarity calculation approaches, word distance calculation approaches and some other statistical approaches to the data of answer analysis, options and question stem and those results are regard as the variables in multiple regression model. This paper use this model to predict weather added new questions are useful. The results of those methods can improve the quality of student exercises. Finally, display the construction and data in knowledge graph.On the basis of the constructed knowledge graph, this paper designs and develops a question selection application. First, introduce the student knowledge graph generative model and design question selection model based on knowledge graph and the wrong-question records. Subsequently, explain key solution methods in the question selection model, including the algorithm of question selection based on knowledge graph and those based on wrong-question records, student knowledge level assessment rules and difficulty level adaptive-updating rules. Then, propose requirements and system design in real teaching scenarios. Finally, display the effect of question selection system and experiment result in real teaching scenarios.Combining education theory, this paper proposes question selection approaches based on knowledge graph and wrong question, and designs and implements the teacher-led question selection system. Experiment results shows that those approaches and this system can help student to improve knowledge level and study efficiency and also can help teachers to master student knowledge level which can effect to find the weak in the course study.