Learning management system with personalized recommendation based on agent BDI

Be aimed at solve the problem of insufficiency of personalized recommendation function in the open and distance learning (ODL) information services, this article puts forward a learning management system with personalized recommendation function (RLMS) based on agent. A learner agent (LA) is set up, which substitutes an online learner, maintains and analyzes learners´ actions and interest. A recommendation agent (RA) is also set up in the system. It recommends the learning resource for the learners, which fits their individual characteristics. Three intentions are set up for RA to express three kinds of recommendation logic. The description of learner´s information model, learner´s interests model, the matchmaking process, the definition of RA´s intentions as well as actions, the recommendation algorithm are also discussed in this paper. Meanwhile, a prototype has been built to verify the validity and efficiency of the algorithm. The prototype system uses JADE as the design platform of agent and uses Lucene to serve retrieving of learning resources. Experiments based on real data showed that this mechanism can recommend learner learning resource properly and effective, and it has sustainable improved efficiency and scalability of collaborative learning for ODL learners.