A Study Of Construction And Application Of Human Resources Post Competency Model Based On Text Mining

In the context of the post-epidemic era,the number of both supply and demand for companies and job seekers has increased.The convenience,safety,low cost and strong interaction of online recruitment platforms have continued to increase the number of users.The amount of online recruitment information data has also exploded.At the same time,recruitment as one of the modules of human resource management,the workload of selecting and hiring personnel is also increasing.In order to help companies improve the talent profile,accurately identify the required talents,help job seekers keep up with the update iterations of knowledge and technology,and improve their own competencies,this article uses big data text mining technology to obtain human resource management positions released by the top online recruitment platform Recruitment information of the past two months,through the method of vectorization,K-means clustering and word cloud diagram in this article,a competency model for human resource management positions is constructed-a four-factor model: workplace skills,business knowledge,strategic contribution,and personal characteristics,And these four factors in turn present a hierarchy from top to bottom and from the inside to the outside,representing the core competence of human resources positions that are increasingly difficult to be detected and cultivated.In order to reduce the time-consuming,labor-intensive and low-value work of selecting resumes in recruitment,improve the efficiency of recruitment and selection,and more accurately conduct training and performance evaluation based on people-post-organization,This article uses the expert evaluation method,based on the human resources job competency model established above as the standard,100 job applicant resumes are scored,and then the logistic regression algorithm and multiple regression analysis method are used to construct a machine learning model for competency prediction.The model found that the factors that have significant influence on the comprehensive competency quality of job applicants include working years,the scale of the company,the school level,whether they are human resources majors,whether they have a human resources manager certificate,and whether they have relevant training experience,whether Factors such as awarding experience and the highest academic qualifications and competence have no significant impact.Finally,four types of indicators including accuracy rate,recall rate,accuracy rate and F1 score are used to comprehensively verify that the machine learning model has a good predictive effect on competency,and predicts whether the candidate's resume is highly competent or low-competent.The accuracy rate on two classification problems can reach more than 70%.The machine learning model of competency prediction can be used in the practical work of enterprises to realize automation,mass screening of resumes,or evaluation of the competence of talents.