Research On College Entrance Examination Voluntary Recommendation System Based On Deep Feature Extraction

College entrance examination voluntary filling is an important part of the college entrance examination,which is of far-reaching significance to candidates and parents.However,in the face of complex and diverse college and professional information,it is difficult for candidates and parents to make the choice that best meets their own needs in a short time.Personalized recommendation can recommend interested information according to user characteristics and preferences.It has been successfully applied in teaching and scientific research fields such as online education and expert screening.For the college entrance examination voluntary filling,this paper studies the personalized college entrance examination voluntary recommendation system.Combined with personal work content and many years of work experience in college entrance examination recruitment,this paper constructs the college entrance examination voluntary recommendation text,extracts the features of the text by using the improved convolution neural network,optimizes it,and designs the prototype system and display.The main work of this paper is as follows:(1)In view of the current situation and problems that the current college entrance examination voluntary recommendation website system only carries out keyword retrieval and cannot understand the semantic features for personalized recommendation,this paper obtains the college information of the official and major platforms and the evaluation data of professional disciplines of colleges and universities of the Ministry of education,and constructs the college entrance examination voluntary recommendation text to train the word vector.The improved convolution neural network is used for semantic feature extraction,which improves the problem that the current mainstream websites and systems only retrieve keywords without semantic information.The extracted features trained by this method have stronger semantics and representation ability.(2)Aiming at the problem that examinees and parents have less understanding of colleges and majors,and the proposed professional intention is more colloquial and life-oriented,this paper makes targeted improvements on the basis of extracting semantic features,and comprehensively optimizes the degree of personalization and comprehensive weight by using content-based recommendation algorithm and the method of integrating cosine similarity and common words,so as to make the effect of personalized recommendation more complete.Finally,this paper studies and designs the personalized college entrance examination volunteer system based on feature extraction.In the word,aiming at the personalized recommendation of college entrance examination volunteers,this paper constructs a relatively complete college entrance examination volunteer text,designs an intelligent algorithm based on convolutional neural network for semantic feature extraction,and makes targeted optimization.Finally,this paper designs and displays the corresponding personalized college entrance examination volunteer recommendation system.