Design And Implementation Of College Entrance Examination Voluntary Recommendation System Based On Hybrid Recommendation

"The college entrance examination is difficult,and it is even more difficult to fill in the report" has always been a significant problem faced by every college entrance examination family.A large number of colleges and majors information,the annual college entrance examination policy reform,and changes in the enrollment plans of various colleges make it difficult to fill in the college entrance examination volunteers.How to extract relevant information in line with the candidates’ situation from such complex and huge information is an urgent problem to be solved.With the rapid development of Internet technology,recommendation technology has also been greatly developed.Aiming at the problem of difficulty in choosing suitable colleges and majors in the college entrance examination volunteer application,this thesis designs and implements a college entrance examination volunteer recommendation system based on mixed recommendations.The main research work of this thesis is as follows:(1)This thesis uses the GM(1,1)prediction model and the Verhulst prediction model in the grey system theory to predict the admission score line,uses the posterior difference method to test the prediction results,and selects the best result as the final admission score line prediction value.This thesis takes the science admission scores of2756 colleges and universities in the Inner Mongolia Autonomous Region from 2017 to 2020 as the input.The test results show that in 97.16% of the colleges and universities,the predicted scores in 2021 and actual admission scores have an error value of 10.(2)This thesis takes the Hollander career interest test and the Myers-Briggs personality classification index as input and completes the professional recommendation for candidates through a hybrid recommendation algorithm based on content recommendation and collaborative user filtering.This thesis uses the normalized depreciation cumulative gain to evaluate the recommendation results.The results show that the recommendation effect of the hybrid recommendation algorithm is better than that of the content-based recommendation algorithm and the user-based collaborative filtering algorithm.When the hybrid recommendation model adopts the TOP-15 method,the overall recommendation works best.(3)In this thesis,the K-means++ algorithm is used to perform cluster analysis on the predicted admission scores of each college and finally realize college recommendations for candidates.This thesis uses the silhouette coefficient to evaluate the clustering effect.The evaluation results show that when the number of clusters is6,the K-means++ algorithm has the best clustering effect.(4)This thesis uses vue,Spring Boot,and other technologies to complete the development of the college entrance examination volunteer recommendation system and finally realizes six functions,including the admission score prediction,professional recommendation,college recommendation,personal information management,data query,and data management.After completing the development work,this thesis systematically tests the college entrance examination volunteer recommendation system and analyzes the test results.At present,the college entrance examination volunteer recommendation system has been tested and operated by an education technology company,which has provided strong support for many candidates to fill in the college entrance examination volunteer work.