Research And Implementation Of Adaptive Learning System Based On Knowledge Graph

The integration of modern technology and education is gradually becoming a new direction to promote the development of China's education construction.However,how to extract effective information from educational big data and so as to meet the learners' adaptive learning needs is an important subject worthy of attention and research.Therefore,this paper takes that is based on the knowledge graph,a visual education tool in the theoretical research,and then transforms the theoretical content into practical application that designs an adaptive learning system.Based on the learners' personalized learning characteristics,the system realizes the learners' adaptive learning and personalized growth.The paper carries out research from the following main aspects:(1)In the background of smart education,with the goal of developing self-adaptive learning,a set of test questions with personalized learning characteristics and learning level is generated.A test set automatic classifying model based on clustering analysis and data mining is proposed,namely TS-AC model.The model is completed in two stages: matching knowledge point label and automatic classifying of test question set.First of all,the test questions are automatically matched with the corresponding knowledge point labels by the text classification technology,which greatly improves the efficiency and quality of matching.Secondly,the data mining is carried out based on the records of learners' answers.By establishing the dynamic data feature classification model and introducing the clustering analysis,the automatic classifying of the test set based on learners' learning features is realized.The test set after classifying has good adaptability and mining ability.(2)After getting the test set which has the characteristics of learners' learning,we will make further analysis and research.Through the collected data of online education platform,multi-dimensional analysis of the personalized characteristics of learners is carried out and presented in the form of label,and clustering algorithm is used to build the learner portrait.In the next step,we use association rule mining method to research and mine the knowledge points in the text questions,and generate the association graph between the knowledge points.According to the topological sorting algorithm,the association graph between knowledge points is transformed into a linear ordered sequence,and the learning path of learners is finally generated.The optimal learning path obtained by the simplified path combined with the learner's portrait construct the knowledge graph of the learners.This method can provide effective reference for the personalized and adaptive learning of the learner.(3)By transforming theory into practice,this paper designs a prototype system of adaptive learning based on knowledge graph.This system will realize the research content of the theory of this paper,and confirm the research significance and results of this paper by practice.Through this system,we can understand the personalized learning characteristics and performance of learners,and adjust the learning behavior and methods according to the learning path generated by the system.Finally,we can achieve the refined progress of education level,and realize the real combination of learning and teaching.