DOI QR코드

DOI QR Code

데이터마이닝 기법을 활용한 대졸자 고용에 미치는 영향요인 분석

The Factors that Affects the Employment Type of The Graduates by Data-mining Approach

  • 투고 : 2012.01.17
  • 심사 : 2012.06.11
  • 발행 : 2012.07.31

초록

데이터마이닝 기법은 대량 데이터로부터 유의미한 정보를 추출하기 위해 고용정보 분야 등 다양한 분야로 폭넓게 활용되고 있다. 대졸자 실업, 여성 재취업, 고령자 취업 등의 불안정한 고용형태가 사회적 문제로 등장함에 따라 고용 안정성을 높이려는 다양한 공공 고용서비스의 노력과 연구가 진행되고 있다. 대졸자의 고용형태(상용직, 무직자, 일용직 등)에 영향을 미치는 주요 요인을 개인생활 환경 또는 학교생활 측면에서 분석한 연구 결과는 효과적인 취업 지원과 대학생의 취업 준비를 지도하는 데에 사용될 수 있다. 이러한 사회 조사 분석은 다양한 요인과 방대한 데이터양으로 인해 일반적 통계적 분석만으로는 한계가 있다. 170여개의 속성들 간의 관계 분석과 2만여 개의 응답데이터를 다루기에는 데이터마이닝 기법이 유용하다. 본 연구는 고용형태에 영향을 미치는 요인을 학교생활 요인, 개인환경 요인, 또는 직장생활 요인으로 구분하고, 의사결정 알고리즘을 통해 연관 관계를 분석하였다. 분석결과 주요 부모님의 소득, 결혼 여부 속성과 같은 개인 환경요인이 개인의 고용형태에 큰 영향을 미치는 것으로 나타났고, 의사결정구조 모델은 87%정확성을 보였다. 학교생활 요인으로는 대학 소재 지역으로 추정되는 학교수준이 주요 요인으로 나타났다.

Data mining technique can be adapted to analysing Employment information in order to discover valuable information out of large data. As the issue employment such as jobless of college graduate, recruitment for women, recruitment for elders etc. became social problem, there are many efforts of various public employment services and studies. The factors that affects the college graduate's employment type (regular, temporary, daily) can be used to guide employment and to prepare employment for college students. In analyzing large number of attributes and the huge amount of data elements, regular statistical methods faces their limitation; therefore, data-mining technique is more suitable for the dataset of about 170 attributes and 20,000 elements. We divide the factors that may affect the employment type into personal factor, school factor, company factor, and experience factor; decision tree algorithm is used to find out the interesting relationship between the attributes of the factors and employment type. Personal factors such as the income of parents and marital status were the most affective factors to the employment type. The learned decision tree was able to classify the employment type with 87% of accuracy. We also assume the level of the school affects the employment type of the graduates.

키워드

참고문헌

  1. S. Kim, and Y. Seo, "Organizational and Occupational Mobilities in Firm and Occupational Internal Labor Markets: The Case of Medical Profession," Korean Sociological Association, Vol 37 No 4, pp. 1-29, 2003.
  2. H. Kim, and I. Jeong, Bi-directional Fuzzy Matching Algorithm," Korean Society for Internet Information, Vol. 12 No. 3, 2011.
  3. J. Ban, K. H. Kim, K. H. Kim, "A Study on Factors Affecting Youth Employee's Labor Mobility and Employment Status Transition," Korean Academy of Social Welfare, Vol 57 No. 3, pp. 73-103, 2005.
  4. H. Song, "A Comparative Analysis of Labor market Structure: Korea and Taiwan," Korean Sociological Association, Vol. 34, pp. 951-979, 2000.
  5. J. Yoon, "Structural Changes in Korean Labor Market and Active Labor Market Policy," Master thesis, Chungnam National Univ., pp. 40-56, 2004.
  6. H. Jang, "A Study on the Programs for vitalization of the "Employment Center" in Ministry of Labor : With the Focus on the 4 "employment center" in Incheon," Master thesis, Incheon Univ., pp. 62-72, 2004.
  7. K. Jeong, and H. Kim, "Quality Status Comparison and Analysis for the Service Development Direction of Domestic Job Information Site," Journal of the Korean Society of Computer and Information, Vol. 13 No. 5, Korean Society of Computer and Information, 2008.
  8. S. Ha, and J. Lee, "Customer Relationship Management Techniques Based on Dynamic Customer Analysis Utilizing Data Mining) ," Korea Intelligent Information System Society, Vol.9 No. 3, 2003.
  9. Agrawal, R. & Srikant, R, "Fast algorithms for mining association rules," The International Conference on Very Large Databases, pp. 487-499, 1994.
  10. Bridges, W. P. & W. J. Villemez, "Employment Relations and the Labor Market: Integrating Institutional and Market Perspectives," American Sociological Review, 56, pp. 748-764, 1991. https://doi.org/10.2307/2096253
  11. J. Han, and M. Kamber, "Data Mining," Elsevier, CA. USA. pp. 279, 2001.
  12. H. Lee, "Analysis of Business Attributes in Information Technology Environments," Journal of Information Processing System, Vol 7, No. 2, pp. 385-396, 2011. https://doi.org/10.3745/JIPS.2011.7.2.385
  13. E. Rosch, "Cognitive Representations of Semantic Categories," Journal of Experimental Psychology, Vol 8, pp. 431-455, 1995.
  14. I. H. Witten, and E. Frank, "Data mining: practical machine learning tools and techniques," Diane Cerra, CA. USA, 2005.