Research On Employment Recommendation Method Of College Graduates Based On User Portrait

Nowadays with more and more college graduates,the number of college graduates this year reached its peak,which makes the difficulty of the graduates to find a job greatly ascend,also have the influence of the outbreak of the new champions league again,most companies hiring quantity are reduced,and make the college graduates want to find a satisfactory job more difficult,The employment situation looks particularly grim.At present,the employment work provided by various colleges and universities,such as double selection meeting,can not improve accurate and personalized employment recommendation and relevant guidance for all graduates.However,the employment websites of major colleges and universities generally only display enterprise recruitment information,and can not make accurate and personalized recommendation according to the different personal conditions of each graduate.This paper proposes an employment recommendation method for college graduates based on user portrait,which can make personalized job recommendation to graduates according to their personal situation,which can be used as a reference for their employment.The main contents of this essay are as follows:1.Aiming at college graduates,The user portrait is constructed from two angles based on the user’s personal basic information and historical behavior information.2.By using the word2 vec natural language processing model,the post information text after text preprocessing is used as the corpus for word vector training;Through the LDA theme model,the post is divided into different themes based on the post information text,and the two are combined to produce the object portrait of the post,that is,the vectorized representation of the post.The method based on TF-IDF and time attenuation is adopted to obtain the weight of users visiting different posts,and the user’s job interest portrait is obtained in combination with the post article portrait with the analysis of user’s information.By calculating the matching degree between users and posts,carry out post recommendation,and further post recommendation according to their job search status by looking for user groups with similar job search interests.When looking for similar user groups,users in the same cluster after clustering with the improved K-Means plus plus algorithm are preferred,so as to reduce the complexity of the operation of the recommendation algorithm and improve the operation efficiency of the recommendation algorithm.3.Finally,the relevant experiments are carried out on the recommendation algorithm proposed in this paper,and the employment recommendation system for college graduates is designed and developed,and the relevant functional modules of the employment recommendation system are completed.The experimental results show that the employment recommendation method based on user portrait for college graduates proposed in this paper can effectively use the personal basic information and historical behavior information of graduates,alleviate the sparsity of user scores,improve the accuracy of algorithm operation,and finally improve the job success rate of college graduates,which is helpful to alleviate the employment pressure of college graduates and improve the pertinence of employment recommendation.