해양환경안전학회지 (Journal of the Korean Society of Marine Environment & Safety)

  • 제26권2호
  • /
  • Pages.121-128
  • /
  • 2020
  • /
  • 1229-3431(pISSN)
  • /
  • 2287-3341(eISSN)
해양환경안전학회 (The Korean Society of Marine Environment and safety)
권호 표지
DOI QR코드

DOI QR Code

텍스트 마이닝 기법을 활용한 설문 문항 개선에 관한 연구

A Study on Questionnaire Improvement using Text Mining

  • 투고 : 2020.01.08
  • 심사 : 2020.04.27
  • 발행 : 2020.04.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

국민의 해양안전문화 수준을 객관적으로 측정하고 해양안전문화 확산을 위한 자료로 활용하고자 2018년에 해양안전문화지수를 개발하였다. 안전문화지수를 산출하는 방법은 안전문화에 영향을 줄 만한 이슈를 포함해야 하고 현 실태를 측정할 수 있는 문항으로 구성되어야 한다. 또한, 사회적·경제적 변화에 따라 지속적인 검증과 보완이 요구된다. 해양 전문가에 의해 설계된 설문 문항이 국민의 관심사와 요구를 잘 반영하고 있는지 확인하기 위해 915명의 해양안전 관련 제안 내용을 분석하였다. 비정형 데이터인 해양안전 제안 내용을 분석하기 위해 텍스트 마이닝 기법을 활용하였으며, 네트워크 분석과 토픽 모델링을 수행하였다. 해양안전 제안을 분석한 결과 '교육', '홍보', '안전수칙', '의식', '전문 인력', '시스템'에 관한 내용이 주를 이루었다. 해양안전 제안 사항이 2019년 설문 문항에 반영되도록 18개의 문항을 수정·보완하였고, 설문 문항의 신뢰도를 분석한 결과 내적 일관성은 0.895로 높게 평가되었으며 전년 대비 향상되었다. 해양 관련 전문가뿐만 아니라 국민의 요구사항까지 반영한 개선된 설문 문항으로 해양안전문화지수를 도출함으로써 해양안전문화 확산을 위한 정책 수립에 더 기여할 것으로 기대된다.

The Marine Safety Culture Index (MSCI) was developed in the year 2018 for objectively assessing the public safety culture levels and for incorporating it as data to spread knowledge regarding the marine safety culture. The method for calculating the safety culture index should include issues that may affect the safety culture and should consist of appropriate attributes for estimating the current status. In addition, continuous verification and supplementation are required for addressing social and economic changes. In this study, to determine whether the questionnaire designed by marine experts reflects the people's interests and needs, we analyzed 915 marine safety proposals. Text mining was employed for analyzing the unstructured data of the marine safety proposals, and network analysis and topic modeling were subsequently performed. Analysis of the marine safety proposals was centered on attributes such as education, public relations, safety rules, awareness, skilled workers, and systems. Eighteen questions were modified and supplemented for reflecting the marine safety proposals, and reliability of the revised questions was analyzed. Furthermore, compared to the previous year, the questionnaire's internal consistency was improved upon and was rated at a high value of 0.895. It is expected that by employing the derived marine safety culture index and incorporating the improved questionnaire that reflects the requirements of marine experts and the people, the improved questionnaire will contribute to the establishment of policies for spreading knowledge regarding the marine safety culture.

키워드

참고문헌

  1. AAA foundation for Traffic Safety(2019), 2018 Traffic Safety Culture Index, http://www.aaafoundation.org/ (Accessed Nov. 2019).
  2. Blei, D. M.(2012), Probabilistic Topic Models, Communications of the ACM, Vol. 55, pp. 77-84. https://doi.org/10.1145/2133806.2133826
  3. Blei, D. M., A. Y. Ng, and M. I. Jordan(2003), Latent Dirichlet Allocation, The Journal of machine Learning research, Vol. 3, pp. 993-1022.
  4. Cronbach, L. J.(1951), Coefficient alpha and the internal structure of tests, Psychometrika, Vol. 16, No. 3, pp. 297-334. https://doi.org/10.1007/BF02310555
  5. Feldman, R. and I. Dagan(1995), Knowledge Discovery in Textual Databases(KDT), Knowledge Discovery, Vol. KDD-95, pp. 112-117.
  6. Griffiths, T. L. and M. Steyvers(2004), Finding Scientific Topics, Proceedings of the National Academy of Sciences, Vol. 101, pp. 5228-5235. https://doi.org/10.1073/pnas.0307752101
  7. Grun, B. and K. Hornik(2011), Topic models: An R Package for Fitting Topic Models, Journal of Statistical Software, Vol. 40, No. 13, pp. 1-30.
  8. Hotho, A., A. Nurnberger, and G. Paass(2005), A Brief Survey of Text Mining, LDV Forum - GLDV Journal for Computational Linguistics and Language Technology, Vol. 20, pp. 19-62.
  9. IAM RoadSmart(2016), Measuring attitudes to driving safety & behaviour_The IAM RoadSmart Safety Culture Index, IAM Safety Culture Report, pp. 2-3.
  10. Kang, S. Y., K. S. Kim, and B. S. Rho(2018), An Analysis of Causes of Marine Incidents at sea Using Big Data Thchnique, Journal of the Korean Society of Marine Environment & safety, Vol. 24, No. 4, pp. 408-414. https://doi.org/10.7837/kosomes.2018.24.4.408
  11. Kim, S. H., Y. J. Lee, J. Y. Shin, and K. Y. Park(2010), Text ming for Economic Analysis, Bank of Korea, Vol. 2019-18, p. 10.
  12. Kim, Y. M.(2013), Study on Improving Safety Cultures by Analysing Behavior Characteristics of Korean Seafarers, Journal of the Korean Society of Marine Environment & safety, Vol. 19, No. 5, pp. 503-510. https://doi.org/10.7837/kosomes.2013.19.5.503
  13. KOSHA(2019), Korea Occupational Safety & Health Agency, Concept of safety culture, http://www.kosha.or.kr/ (Accessed Dec. 2019).
  14. Lee, J. S., B. K. Lee, and I. S. Cho(2019), Text Mining Anaysis Technique on ECDIS Accident Report, Journal of the Korean Society of Marine Environment & safety, Vol. 25, No. 4, pp. 405-412. https://doi.org/10.7837/kosomes.2019.25.4.405
  15. MOLIT(2019), Ministry of Land, Infrastructure and Transport, Transport Culture Index, http://www.molit.go.kr/ (Accessed Dec. 2019).
  16. Paek, Y. J., C. H. Jung, and D. G. Yoon(2018), The Development of Marine Safety Culture Index, Journal of Korean Maritime Police Science, Vol. 8. No. 4, pp. 79-106.
  17. Park, J. H. and M. Song(2013), A Study on the Research Trends in Library & Information Science in Korea using Topic Modeling, Journal of the Korean Society for Information Management, Vol. 30, No. 1, pp. 7-32. https://doi.org/10.3743/KOSIM.2013.30.1.007
  18. Salton, G. and M. J. McGill(1983), Introduction to Modern Information Retrieval, McGraw-Hill, New York.
  19. Seo, S. H.(2016), Fintech Trend Analysis using Topic Modeling of BM Patents, Graduate School of Seoul National University of Science and Technology.