Personalized Collaborative Learning System Students' Model Construction And Application

The rapid development of computer,network and information technology have promoted the popularity of online education,and network-based collaborative learning as a new way of education technology have concerned to many researchers.The current online education is still not a very good platform for students,which could provide suitable studying strategies to fulfill study task effectively.Considering this situation,the personalized collaborative learning system (PCLS) combines individual differences and cognitive abilities of students,provides personalized teaching means,so as to enhance the learning efficiency of the student.As the core of the PCLS-Student Model is the basis and prerequisite of personalized teaching.Therefore,the establishment of a reasonably available Student Model,which will promote the implementation of the intelligent and personalized teaching for PCLS,is very important.Beijing Municipal Higher Education approved the teaching reform project "to create discrete mathematics teaching aids platforms,improve the quality of undergraduate teaching".In this paper,base on this project, It studied the student model for the PCLS thoroughly,and constructed the student model based on the second-stage fuzzy comprehensive evaluation with a combination of the method of Bayesian network.Furthermore,It designed and developed the prototype system of the PCLS based on J2EE platform.The main work in this paper can be described as follows:1.Based on design method of collaborative learning systems,it researched and realized the personalized collaborative learning system for "Discrete Mathematics" curriculum in this paper;2.For organization and implementation of "Discrete Mathematics" curriculum teaching knowledge base,this curriculum was divided into 664 atomic knowledge points,and then organized them through the orderly relationship of them,constituted directed acyclic graphs,and divided grasping that the students counted to knowledge points into four levels;3.A method of combination of fuzzy comprehensive evaluation based on the second floor and Bayesian network to build the student model is presented in this paper,and the individual information of students is comprehensive analyzed and evaluated more comprehensively and completely.These are setting up the mainstream student models and modeling method based on in-depth researching at home and abroad,and according to the cognitive learning theory and construct doctrine thought,it evaluated the test scores of students based on the fuzzy comprehensive evaluation,and evaluated the individual information of students considered based on the second-stage fuzzy comprehensive evaluation.Furthermore,prior probability of each knowledge point is determined by the directed acyclic graph based on knowledge points,and Bayesian Network based on knowledge points is built.Finally,studying behavior of students is reasoned by the strong ability of Bayesian Network,and students are guided to achieve the studying tasks by the appropriate teaching strategies.Combination of the two methods is complementary in the process of reasoning,and has received good results when this system work was working.4.The prototype system of the PCLS is designed and implemented based on "Discrete Mathematics" curriculum.This system is developed from actual demand,and base on the J2EE platform and the method of the plug-in software development,adopt the use of Web Service technology,and has also succeeded the decoupling of controller layer and Business Logic layer, so that the whole system has the characteristics of distribution and has a very good flexibility and scalability.