Research And Implementation Of Test Recommendation Based On Collaborative Filtering And Intelligent Analysis System

Recently the rapid development of cloud computing,big data,Internet of Things and artificial intelligence has promoted the growth of smart education and digital campus.It brings new opportunities to solve the bottleneck problem of educational informationization.At present,examination is still an important way to evaluate students' performance and teachers' teaching achievements.Therefore,more and more attention has been paid to the intelligent generation of test papers,the analysis of generation quality,and the mining of valuable information behind the examination results.At the same time,the test recommendation technology has also been widely concerned by educators.As a common algorithm of recommendation system,collaborative filtering is worthy of in-depth study in the education field.Therefore,it is an urgent problem for modern teaching management to study the method of generating intelligent test papers,the improved collaborative filtering algorithm for test question recommendation,and the mining of potential valuable information behind examination results.This paper first studies the background and significance of the subject.And it analyzes and studies the related work on question recommendation and intelligent analysis of examination results.Then it investigates the improvement of collaborative filtering algorithm and implements the Office plug-in.The details are as follows:(1)The recommendation algorithm for student-side is improved based on the characteristic of education field.The balance factors such as timeliness,number of answers and mastery rate on knowledge points are introduced combined with students' historical records.The traditional collaborative filtering algorithm is improved on similarity calculation and rating prediction.The experimental results show that the recommendation results are improved effectively.(2)Based on Office customized plug-in related technology,the solution is researched in generating test questions and test papers intelligently in multiple ways,which docks with the standard QTI format and general test platform Moodle.(3)Designed and implemented Excel plug-in for intelligent analysis of test results to perform education classical measurement theoretical analysis of students' test scores and multi-dimensional,in-depth mining of test answer results,which helps teachers improve teaching strategies and students improve learning methods and ideas.