The Research and Construction of Network Adaptive Learning System Based on Multi-Agent
|School||Tianjin Normal University|
|Keywords||Adaptive Learning Agent Student Model Bayesian Nets|
With the popularization and application of the network technology in all fields, the distance instruction based on network has already become a research subject payed much more attention in the education field.Through in-depth investigation and analysis, as to the shortage of some network tutoring systems, such as the lack of intelligence, weakly interactive function, faulty navigation, we proposed an idea to better it in this paper.Regarding adaptive learning theory as guidance, we introduced Multi-Agent technology and integrated the maturity technology of the intelligent tutoring system, then constructd network adaptive learning system based on Multi-Agent in the idea.Finally we realized a corresponding network tutoring system, and it has made better result in such aspects as individuation, interaction, adaptability.During the process of designing and realizing network adaptive learning system based on Multi-Agent, the research work of the following several respects has been mainly done in this paper: ￤e constructed B/A/S three layers structure in system design, namely adaptive function display layer ,Multi-Agent layer and data resource layer.The three layers structure incarnates guidance of adaptive study theory and principle of introducing Multi-Agent technology and integrating the maturity technology of the intelligent tutoring system in the stand-alone instance.(2)We constructed student model with Bayesian Nets and Granularity Hierarchies method.They provide basis of strategy of test and student’s adaptive learning.(3)We put forward a Multi-Agent system regarding finishing a adaptive study task as the goal and realized the simple communication function based on FIPA-ACL language between Agents.The system can estimate knowledge characteristic and learning style of student by test.According to knowledge characteristic and learning style of student, thesystem can provide adaptive learning space.The system realized adaptive learning on the whole.And the embodiment of adaptability is: adaptive content presentation, adaptive navigation, and adaptive help.Among them, adaptive help includes: adaptive companion help, adaptive teacher’s guide, knowledge query based on keywords.In conclusion, this paper is only a summary of research and exploration about network teaching.Improving intelligence, interaction, navigation of network teaching, humanizing network teaching, and implementing teaching students in accordance of their aptitude still need constant further study and development.