Research And Application Of University Students Employment Recommendation System Based On Deep Learning

With the increasing number of graduated students,the overall employment situation of college graduates has become increasingly severe since 2011.At the same time,It is a great challenge between college students expectation and employer qualification,which lead to a large number of college students can’t find suitable jobs,but the employer cannot find the suitable employee.This phenomenon is mainly due to the huge information gap between graduates’ employment and employers.Therefore,There is a huge gulf between the current colleges and employee,and the information bridge should be built.Thus,how to use deep learning theory to analyze student information reasonable analysis to build a highly reliable recommendation model to provide graduates with a reliable employment orientation reference can effectively alleviate the current employment tension.A feature combination hybrid recommendation algorithm based on deep learning is proposed in this paper,and it is experimentally verified in the field of employment recommendation for college graduates.With the recommendation model as the core,a deep learning-based employment recommendation system for college students is designed and implemented to provide college graduates with an employment recommendation system.Provide accurate,college,and convenient employment recommendation services.The research and application are mainly carried out from the following three aspects.(1)the three classic algorithms,which are collaborative filtering,random forest and BP neural network recommendation algorithm,in deep learning are compared,analyzed and summarized.Then taking the basic information of graduates of Harbin Engineering University and the data set of employers as input samples,the application effects of the three models are verified respectively,and the evaluation indicators are compared to draw conclusions.(2)Aiming at the problems of low recommendation accuracy,limited feature representation ability,and sparse data of the three traditional algorithms in deep learning,a deep learning-based feature combination hybrid recommendation model is proposed,which is an improved recommendation model under traditional recommendation algorithms.,the algorithm is implemented under the same input data set.The experimental results show that the recommendation effect of the analysis feature combination hybrid recommendation model under this dataset is due to the classical model,which has strong learning ability and high scalability.The improved model can improve the accuracy of recommendation and alleviate the existing recommendation to a certain extent.The problem of poor representation of data features in the model and data sparsity.(3)According to the actual needs of graduates,expand the data set,and develop an employment recommendation system suitable for college graduates,the improved feature combination hybrid recommendation model is applied to the employment of college students.The system has modules such as registration and login,student preference collection,data processing,and recommendation result display.It can realize that users can search and view information of employers in the system,and can also realize the function of personalized recommendation to employers,providing a graduate employment recommendation platform with strong guidance and reference for college graduates when they are looking for a job after graduation.The research results shows that the system is strong practicability,good employment experience and high employer satisfaction.And it is of great significance to improve the quality of employment.