Design And Implementation Of Learning Resource Recommendation Platform By Integrating User Profile

With the rapid development of the Internet in recent years,people pay more and more atten-tion to online education.In particular,the sudden outbreak of COVID-19 has highlighted the importance of online education.Online education has innovated the traditional learning model.Learners can access various learning resources through the Internet.With the thriv-ing of online education,the number of learning resources on the Internet has also increased,thus leading to the overloading of learning resources.Learners find it difficult to choose suitable learning resources.Therefore,it is of great significance to build a personalized rec-ommendation platform for learners to improve the efficiency of online learning.Due to the particularity of learning resources,general recommendation algorithms have problems such as low performance and low accuracy in learning resource recommendation,which leads to poor user experience.Moreover,the current online education platforms do not build user profile to recommend learning resources.Based on the above background,this thesis expounds the necessity of personalized recom-mendation of learning resources and studies the researches carried out by domestic and for-eign scholars in the field of learning resource recommendation.Then,this thesis analyses the requirements,architecture and algorithms of the learning resource recommendation plat-form in detail.As for the algorithm,this thesis proposes a learning resource recommendation algorithm by integrating user profile.This thesis takes college students as the research ob-ject,and constructs user profile by mining users’ personal attributes,historical behaviors and professional information.According to the characteristics of user groups and learning resources,this thesis implements and improves the recommendation algorithms from three perspectives.Then calculate the compatibility between the user and the learning resources by integrating user profile with recommendation algorithms.The main work of the thesis is as follows:1.Use web crawlers to grab user data and course data from an online education website to form an experimental data set.Construct user profile by mining users’ personal attributes,historical behaviors,and professional information,and use the vector space composed of[(tag: weight)] to represent user profile,which is used to identify the user’s knowledge struc-ture and interest preferences.2.Implement and improve the recommendation algorithms from three perspectives,from which we can get a more accurate candidate set.In this thesis,the improvement of the user based collaborative filtering algorithm is to increase the weight of main courses in the similarity calculation,the improvement of the apriori algorithm is to use the hash table to solve the problem of frequently scanning the database in apriori algorithm.And analyze the user’s professional information,generate recommendations based on the professional char-acteristics.3.Implement the calculation method of the compatibility between the user profile and the learning resource,and integrate user profile with recommendation algorithm.Use recom-mendation algorithm to generate a candidate set,then calculate the compatibility between the user profile and the items in the candidate set,which will generate the top-n items with the highest compatibility.Compared with calculating the compatibility of all items and users,this method reduces the amount of calculation and improves the accuracy of recommenda-tion.4.Design and implement a learning resource recommendation platform based on Django.Analyze platform’s requirements,architecture and functional modules,and embed the learn-ing resource recommendation algorithm that integrates user profile with recommendation algorithm into the platform.Then carry out function testing and performance testing.With this learning resource recommendation platform,learners can obtain personalized learning resource recommendation services,which effectively improves the learning efficiency of learners.