Research On The Recommendation Method Of Preparing Lessons Resources Based On Collaborative Filtering Algorithm

With the education information developed rapidly, the number of educational resources appear explosive growth, in lesson preparation system in the phenomenon, lesson preparation system is no longer as before learning a small number of resources, users can not get adequate resources. But with the explosive growth of resources, the recommended technology will be applied to the lesson preparation system, to provide users with resources, reduce the user's search time. However, there are two problems:first, in the classification of resources and the extraction of user interest points is not enough detail. Lead to the lack of personalized recommendation resources. Second, the recommended resource quality is low, not in terms of resource quality screening.Collaborative filtering recommendation algorithm uses swarm intelligence to explore the potential interests of users, and does not need to consider the type of resource projects, with a good field of applicability and recommendation of novelty. At home and abroad, collaborative filtering recommendation algorithm has been widely used in many fields, and has achieved remarkable results. Therefore, this paper selects collaborative filtering recommendation technology to solve the above problems, to provide users with high quality and detailed knowledge of the specific points of lesson preparation resources.The main work of this paper are:1) preparation of resource modeling, according to the knowledge resources of the lesson to classify the resources to push resources to target specific knowledge, so, in terms of resources to ensure that resources are recommended to meticulous knowledge of.2) According to the new user to fill in information explicit modeling, using content filtering algorithm is recommended for new users, collaborative filtering algorithm to solve the problem of cold start.3) to obtain explicit information implicit user behavior information and fill in user, create a user model to reflect the user is interested at this stage teacher knowledge points.4) For the resources in the system, the establishment of resource evaluation system to ensure the quality of recommended resources.5) Design and implementation of lesson preparation resources recommendation system, tested, recommended resources considering whether the user wants specific knowledge resources and quality of the merits of the recommended resources, concluded that about 90% of users recommended the first three lesson resources are relatively satisfied with the results when the recommended increasing the previous ten recommendations results, not comment due to the increase of the number of user satisfaction decreased slightly.