텍스트 마이닝을 활용한 대학생 중소기업 취업 주요 이슈 분석

Abstract

This study conducted text mining analysis using news data to objectively evaluate the effects of various policies, such as support for college student and small business employment, implemented during the tenures of past presidents. After collecting a total of 13,939 news articles, preprocessing such as removal of stop words was performed, and keyword analysis and correlation analysis were conducted. The news data was collected from 'Big Kinds' of the Korea Press Foundation. The main results analyzed are as follows: initially, a high-level analysis of the entire period showed that 'jobs' were the most frequently mentioned, followed by 'large companies', 'Seoul', 'workers', and 'job seekers'. Secondly, the yearly word cloud analysis results showed that President Moon Jae-in was associated with Covid-19, manufacturing, subsidies, minimum wage, in that order. President Park Geun-hye was associated with 'workers', youth, subsidies, manufacturing, in that order. President Lee Myung-bak was associated with 'unemployed', manufacturing, vocational high schools (Meister schools), college students, in that order. Thirdly, the TF-IDF analysis revealed distinct differences according to the timing of presidential changes. Additionally, among keywords related to government policies, terms such as National Employment Support System, Tomorrow Learning Credit, and Employment Encouragement Allowance emerged. Through this, it was possible to confirm that these support programs are mentioned in relation to college student employment in small businesses among various government support initiatives.