Research And Application Of Course Recommendation Algorithm In Wisdom Education Platform

With the implementation of a series of major national strategies,the vigorous development of the new economy represented by new technologies,models,industries,and formats has gradually put forward higher requirements for scientific and technological talents.At the same time,in order to explore a new Chinese model that leads the global engineering education,actively promoting the construction of new engineering subjects has become the focus of the Ministry of Education.Based on the educational information paradigm of modern engineering theory,big data analysis,and artificial intelligence of the new engineering system,colleges and universities form a smart education platform that is student-centered,ability-oriented,and continuous iteration of qualified evaluation and course quality.The smart education platform is a new way of education communication accompanied by the continuous improvement and development of the Internet and the digitization and informatization of education.On the one hand,it not only brings great convenience to students,but also provides new learning methods;on the other hand,it also proposes solutions to the phenomenon of "information overload" caused by the rapid and incremental growth of learning resources;Students who do not have a basic understanding of learning courses also bring more choices in choosing courses and their learning paths.The main research direction of this topic is to improve the measurement method of user similarity,and refine the user's behavior and combine the user's learning goals and other factors to improve the accuracy of the recommendation.First,the system forms the domain knowledge system of the university into a knowledge model structure,mines user interest items through user behavior,and updates the interest model in conjunction with the user memory curve model to adapt to the attenuation of user interest;second,extracts and represents the resources in the system Features,and predict the user's intention to score video resources through user behavior to reflect the advantages and disadvantages of the video;then use the user-based collaborative filtering algorithm as a bridge to provide targeted video learning for the user's portrait and learning path Resources;Finally,conduct system tests by simulating real user behavior.