Design Of University Employment Information Service System Based On Big Data

In recent years,the overall number of college graduates has maintained a growing trend.The problem,how to help graduates cope with the increasingly severe employment situation,has become the focus of the current employment service departments.With the development of Internet information technology,job hunters can get job information by browsing recruitment websites.However,there is a lot of false and disordered information in online recruitment,so it is difficult for graduates to obtain rich and comprehensive employment information and personalized recommendation service,which reduces the efficiency of job hunting.In addition,unstructured online recruitment data contains a large number of implicit demand knowledge,which can help the university adjust the teaching plan.But the implicit demand knowledge is difficult to be obtained by job hunters or university teaching managers.In order to solve the above problem,a university employment information service system based on big data is built in this paper.The system of university employment information service system based on the Spark platform processes the job data collected from the job recruitment websites to realize the functions of job data visualization,personalized job recommendation,employment hot spot analysis,association rule mining,etc.This system uses Scrapy web crawler technology to obtain job data from the Internet and save it to Mongo DB.The algorithm of TF-IDF is applied to extract the features of the job data,and the algorithm of K-means is used to cluster the data offline.Then,personal preferences are obtained by analyzing the user’s resume and user behavior log.On this basis,a hybrid offline recommendation strategy of collaborative filtering based on content and model is designed.At the same time,the system designs a real-time job recommendation strategy based on Spark Streaming technology.Offline recommendation and real-time recommendation constitute the personalized job recommendation function of the system.In addition,the LDA Algorithm is applied to analyze popular skills in the job market,and the FP-growth algorithm is used to mine association rules between different data items of job data,so that realizes multi-angle analysis of the job data.This system can provide personalized job recommendation service for job hunters,and provide a powerful help to reduce the stress of the employment problems for college graduates.At the same time,the demand knowledge is got by analyzing the job data and is displayed in this system with diverse visual charts,which can help colleges and universities obtain the feedback of the employment market in time.The feedback can guide the teaching work,track the skills needed by the employment markets,optimize the teaching subjects,and have practical significance for colleges and universities and relevant educational institutions to carry out teaching adjustment and teaching research.