Analysis And Application Of Online Recruitment Information Based On Data Mining Technology

With the rise of big data,data mining technology has also developed rapidly,becoming a hot topic for scholars to learn and research.It is inevitable to generate massive data in the network era.How to use data for analysis and mining to obtain valuable information has become an effective way to gain a competitive advantage.Currently,recruitment websites are emerging one after another,accumulating a large amount of recruitment information.The large and chaotic data makes it difficult for both job seekers and recruiters to quickly match,leading to the phenomenon of companies lacking talent and talent not working.Based on data mining technology,the dissertation crawled,analyzed and modeled the online recruitment information,aiding for rapid matching between recruiters and job seekers.The dataset in this dissertation comes from 51 job,a large domestic recruitment website.According to the characteristics of the website,taking the recruitment information of Python related position as an example,we designed a web crawler program,and finally collected 8042 recruitment information,including 10 fields.Then the obtained data sets were preprocessed respectively,the structured data is cleaned and transformed,the unstructured text data was segmented through the Jieba library,and then the stop words library was updated according to the actual situation to delete the stop words;Then we analyzed the preprocessed dataset,which mainly from three aspects: the characteristics of Python job recruitment companies,job requirements,and job text descriptions.Finally,the SMOTE algorithm was used to balance the salary sample,data transformation was carried out for the five characteristics of the region,industry,experience requirements,educational level,and company size.One-Hot encoding was used in fifteen characteristics extracted from the job description,and KNN classification model,Decision Tree classification model,and Random Forest classification model were established,model evaluation was conducted through the prediction accuracy of the three models,and the accuracy of the Random Forest classification model was relatively high,Using it can help job seekers locate companies that meet their salary requirements,and also enable recruitment companies to understand the level of their company’s salary,thereby adjusting company strategies and completing talent recruitment faster.