Design And Implementation Of Lesson Preparation Resource Recommendation System Based On Django Framework

At present,the development of artificial intelligence empowers the field of education.From the perspective of serving teachers,it has the advantages of reducing burdens,helping teachers pay more attention to the development of student personality and improving work efficiency.For example,the lesson preparation technology of artificial intelligence can meet the needs of teachers for personalized lesson preparation,it can push different teaching resources for teachers according to their different needs,and provide intelligent lesson preparation services.The study found that there are two main problems in the existing lesson preparation platforms.Firstly,the existing lesson preparation platforms are not facing all teachers.They are mainly for K12 teachers,but ignore the need of college teachers for lesson preparation.Secondly,the existing lesson preparation platforms fail to realize personalized recommendation of teaching resources.With the increasing amount of data,it becomes more and more difficult for us to find the right resources among the massive resources,and users also hope that the system can make specific recommendations according to the different needs of each person.Based on the above analysis,this article proposes a lesson resource recommendation system based on the Django framework.This system is for college teachers.Since there are many majors in the college and the classification is meticulous,we take the computer science as an example.The design idea is as follows.Crawling technology is used to crawl computer-related teaching resources from the network,and a Django framework is used to build a lesson preparation platform.Secondly,a labelbased recommendation algorithm and an association rule-based recommendation algorithm are integrated into the platform to implement personalized recommendation of teaching resources.The purpose of this is to use the tag to help us solve the cold start problem in the recommendation system.New users can choose the interest tags during registration to let the system quickly understand the users' needs,so as to avoid the problem that the system cannot recommend suitable resources for new users.At the same time,in order to make the recommendation results more accurate,in the tag recommendation,the TF-IDF algorithm is used to calculate the weight of the tags.Furthermore,in order to feed back more rich recommendation results to users,association rules are incorporated,which can analyze the links between tags,so as to dig out the potential interest tags of users and recommend the resources under the potential interest tags to users.This system uses Python and MySQL database to develop.The development environment is PyCharm.Finally,we use dataset and user simulation to test the system.