Data Mining Techniques Application And Research In Human Resource Market

In recent years,more and more recruiters choose the information publishing platform provided by the personnel market to find suitable jobs.However,the amount of information of recruitment and resume is growing exponential from the Internet,and the manual way to search jobs and talents information is inefficient,which is unable to meet the growing needs of the business.This paper focuses on the application of data mining technology in the personnel market,trying to dig out the potential relationship between candidates and jobs from the huge amounts of data,and realize personalized job recommendation for the job seekers preliminarily.In this paper,chapter 2 analyzes the operation features of the personnel market and data mining technology.On the basis of chapter 2,chapter 3 makes a comparative study between C4.5 algorithm and ID3 algorithm,and the final choice is C4.5.chapter 4,C4.5 algorithm is employed for data mining,and the regular pattern of classification and elicits the classification rules,which is based on the decision tree evaluation model of the personnel market can be found out by using talent information database as the data source from a city level personnel market.It uses the data which is processed by data extraction and data preprocessing as the input of C4.5 algorithm,and it expresses the result decision tree and classification rules as the preference matrix.In chapter 5,a newly talent recommendation method based on the results of the data mining is given out.With rigid conditions filtration,non-rigid conditions similarity calculation and preference calculation,this method utilizes various factors to build a recommended list and provide valuable decision information for the applicants.In chapter 5,an experiment has been designed to compare the performance of three typical recommendation methods,including data mining basis,content filtration basis and collaborative filtration basis.The result of the experiment shows that the proposed method has better recommendation effect compared to the traditional methods: when the recommendation list has more than 10 items,the precision increases about 7%;when the recommendation list has more than 20 items,the recall increases about 6%-9%.The experimental results show that the data mining basis recommendation method is more suitable for the human resource manageme nt needs in the big data world.