Research And Implementation Of Online Learning System Based On Knowledge Tracking

With the constant progress of computer technology and communication methods,different from traditional face-to-face teaching,teachers’ teaching modes and students’ learning methods have undergone earth-shaking changes.The sudden outbreak of the new crown epidemic at the end of 2019,the development of multimedia technology and Internet technology,all of which have prompted the method of online learning to be more recognized by everyone.Online learning systems have become more popular amid a national push to minimize contact.Because of its properties such as mountains and seas of learning resources,varied learning techniques,flexibility,and freedom from time and location constraints,the online learning system has steadily become an integral aspect of online education.Today,personalization has become the core of online learning systems.To provide personalized education for each user,it is necessary to understand the user’s knowledge.Because most systems lack the analysis of data related to the relationship between textbook knowledge points and students’ learning progress,the recommendation effect cannot meet the individual needs of users,and knowledge tracking is to model based on the relevant data of users’ historical answers to predict users’ knowledge of knowledge control of the situation.How to accurately understand the user’s knowledge mastery has become a hot issue to be solved now.Most researchers focus on the model itself,ignoring the explicit and implicit relationship between knowledge points and knowledge points,and directly based on the acquired knowledge mastery.Recommendation,resulting in the recommendation result is either the knowledge that the user has mastered or the knowledge that the user has not mastered at all.Based on this,this paper proposes a knowledge recommendation model based on knowledge tracking to recommend personalized knowledge to users.First,a comprehensive knowledge relationship graph is obtained by fusing the relationship between the knowledge points marked by experts and the potential relationship between knowledge points obtained from the user’s answering behavior data;In knowledge tracking based on gated graph neural network,information aggregation of each node,state update and output prediction are performed to obtain the user’s mastery of each knowledge point.Then,the weight is determined according to the out-degree and in-degree of the knowledge point in the knowledge relationship graph.The larger the out-degree,the more basic the knowledge point is,and the larger the in-degree,the more the knowledge point needs to be combined with other knowledge points to be mastered.The user’s mastery of each knowledge point is combined to obtain the recommendation score of each knowledge point,and the recommendation sequence is obtained according to the level of the recommendation score.Finally,through the experimental verification on the two datasets of Assistments2009-2010 and KDD Cup2010,the model proposed in this paper has improved in terms of AUC and accuracy compared with other knowledge tracking models,proving the feasibility of the model in this paper and effectiveness.This paper develops an online learning system based on knowledge tracking and knowledge recommendation model,which uses users to be divided into two roles:teacher and student.First,fully investigate the relevant needs of teachers and students for system functions.Second,based on the survey results,a complete demand analysis of the system is conducted to offer a meaningful foundation for the system’s overall framework design.Thirdly,carry out the overall design of the system,and carry out detailed design of each functional module and data storage design.Then,according to the detailed design of each functional module,the system is implemented step by step and the model proposed in this paper is applied in this system.Finally,through functional testing and performance testing,the system can meet the needs of students to answer questions online,view test records,modify personal information,and review wrong questions,as well as teachers’ topic management,student management,test paper management,and task release functions need.