Design And Implementation Of Intelligent Recommendation System For Campus Recruitment

Nowadays,with the continuous growing of college graduates every year and increasing employment pressure of college students,it is especially important for every graduate to find a suitable job for a limited period.Combining the previous trend of employment,this paper accomplished the design and realization of intelligent recommendation system for campus recruitment according to the problem of position recommendation for current college graduates.To begin with,this paper summarizes the technical background,research direction and current situation at home and abroad of the recommendation system.Then it introduces the key technology related to the recommendation system and analyzes the current mainstream network recommendation system and also dissects the merits and demerits of the recommendation algorithm commonly used at the present stage.Two innovation points are proposed in this paper on the basis of the conclusion of results.On the one hand,aiming at the problem of poor precision of web page extraction in the current web page extraction field,combined with label path coverage and multiple text features,CETD-TPF,the text extraction algorithm based on label path coverage and multi-text features is proposed and implemented,which has improved the extraction accuracy of web page text.On the other hand,aiming at the sparsity of the matrix and cold start of the recommendation algorithm,this paper adds the time factor on the basis of the traditional collaborative filtering recommendation algorithm.Meanwhile,it also combines with the users clustering analysis of feature similarity and puts forward that considering the collaborative filtering recommendation algorithm based on clustering and user comprehensive interest as the recommendation engine of the intelligent recommendation system,which has completed the design and implementation of the intelligent recommendation system for campus recruitment.By this way,the recommendation system would become more intelligent while avoiding the shortcomings of the traditional recommendation system,so that the recommendation result is more accurate.Experiments show that CETD-TPF,the text extraction algorithm based on label path coverage and multi-text features proposed in this paper,can effectively improve the extraction accuracy of web page text,and the collaborative filtering recommendation algorithm based on clustering and user comprehensive interest can effectively solve the problem of the cold start of project and data sparseness,which has improved the recommendation accuracy and achieved good recommendation results.