The Journal of Korean Institute of Information Technology (한국정보기술학회논문지)

Korean Institute of Information Technology (한국정보기술학회)
권호 표지
DOI QR코드

DOI QR Code

Linguistic Features Extraction for Analysis of Great Natural Disasters Damage State

대형 자연재난 피해상황 분석을 위한 어휘 자질 추출

  • 강주연 (전북대학교 기록관리학과) ;
  • 장효정 (전북대학교 기록관리학과) ;
  • 오효정 (전북대학교 기록관리학과, 문화융복함 카이빙 연구소)
  • Received : 2017.07.07
  • Accepted : 2017.09.15
  • Published : 2017.09.30
⟨ Previous Next ⟩

Abstract

This study aims to propose a method for timely analysis of great natural disasters damage state based on contents because current damage analysis system has problems with time and human recourses in spite of perpetual damage by natural disasters. Linguistic patterns are extracted from the sentences which describe damage state in news articles reporting 4 types of great natural disasters - typhoon, heavy-rain, major snow, and earthquake. After that, the damage targets and types are analyzed by characteristics of predicate-argument structure drawn from the linguistic features and definitions of Korean Standard Dictionary. As the result, even though damage targets and types are different by disasters types, they would be similar when the causes to lead to damages are same. The result of this study helps improving accuracy of damage analysis for minimizing the omission of disasters damage as well as an alternative method for minimizing the loss of time and human recourse to rapidly aggregate the damage states.

본 연구는 재해 현장의 피해조사 체계가 지니는 시간과 인력의 한계점을 보완하기 위해 재난 관련 보도문의 내용분석을 통해 피해상황을 신속하게 파악할 수 있는 방안을 마련하고자 한다. 이를 위해 태풍, 호우, 대설, 지진 4가지 유형의 대형 자연재난 상황을 보도한 신문기사에서 피해상황을 전달하는 문장을 추출한 후 어휘 자질의 특성을 분석하였다. 특히 문장 내'술어-논항(Predicate-argument)'구조의 특징과 표준국어대사전을 참고한 의미 분석을 통해 재난 유형별 피해대상과 유형을 파악하였다. 어휘 패턴 분석 결과, 재난 유형에 따라 피해대상과 유형은 상이하였으나 피해를 유발하는 원인이 비슷한 재난의 경우에는 피해대상과 그 유형이 상당부분 일치하는 것으로 나타났다. 본 연구의 결과는 재난 피해 집계의 누락을 최소화할 뿐만 아니라 피해상황을 신속하게 집계하기 위한 시간과 인력에 한계를 지니는 현재의 피해 집계 방식의 대안으로서 운용될 수 있을 것이다.

Keywords

Acknowledgement

Supported by : 한국연구재단

References

  1. Ministry of Public Safety and Security, Policy Description, http://www.mpss.go.kr/home/policy/policy/policydesc/?boardId=bbs_0000000000000047&mode=view&cntId=1696&category=&pageIdx=&searchCondition=&searchKeyword=, [Accessed: Jan. 05, 2017].
  2. S. I. Cho and J. A. Bae, "A Study on Urban Disaster Management Using Network Analysis : Focus on Landslides on Umyeon Mountain", Journal of Institute for Social Sciences, Vol. 27, No. 1, pp. 291-309, Jan. 2016.
  3. W. J. Choi and J. W. Cho, "Development of Work Process on Natural Disaster Damage Investigation using Smartphones", Korean Society of Disaster & Security, Vol. 5, No. 2, pp. 1-10, Dec. 2012.
  4. Y. J. Choi and S. H. Kweon, "A Semantic Network Analysis of the Newspaper Articles on Big data", Journal of Cybercommunication Academic Society, Vol. 31, No. 1, pp. 241-286, Mar. 2014.
  5. J. W. Cho and W. J. Choi, "Efficiency Analysis of Natural Disaster Damage Investigation System using Smartphones", Korean Society Of Hazard Mitigation, Vol. 14, No. 4, pp. 163-170, Aug. 2014.
  6. J. U. Lee, G. G. Yang, S. S. Byun, and D. W. Park, "Improving the Investigation System of Natural Disaster Damage", Crisisonomy, Vol. 5, No. 1, pp. 73-84, Jun. 2009.
  7. H. J. Oh, B. H. Yun, C. J. Yoo, and Y. Kim, "Trend Analysis using Spatial-Temporal Visualization of Event Information based on Social Media", Korean Society for internet information, Vol. 15, No. 6, pp. 65-75, Dec. 2014.
  8. H. J. Oh, B. H. Yun, and C. Y. Kim, "Linguistic Features Discrimination for Social Issue Risk Classification", KIPS Transactions on Software and Data Engineering, Vol. 5, No. 11, pp. 541-548, Oct. 2016. https://doi.org/10.3745/KTSDE.2016.5.11.541
  9. Ministry of Public Safety and Security, 2015 Annals of disaster, http://www.mpss.go.kr/home/policy/statistics/statisticsData/?boardId=bbs_0000000000000042& cntId=27& m ode=view& category=, [Accessed: Jan. 05, 2017].
  10. National Library of Korea, Online Archiving and Searching Internet Sources(OASIS), http://www.oasis.go.kr/collection/mainDisa.do, [Accessed: Jan. 05, 2017].
  11. BigKinds, https://www.bigkinds.or.kr/, [Accessed: Jan. 05, 2017].
  12. S. J. Lim, C. K. Lee, and D. Y. Ra, "Dependency -based semantic role labeling using sequence labeling with a structural SVM", Pattern Recognition Letters, Vol. 34, No. 6, pp. 696-702, Apr. 2013. https://doi.org/10.1016/j.patrec.2013.01.022
  13. S. Raheja, R. Dhadich, and S. Rajpal, "An Alternative Interpretation of Lingustic Variables as Lingstic Finite Automata", Formal Languages and Automata Theory, International Journal of Computer Science & Issues, arXiv preprint arXiv:1112.3435, Aug. 2011.
  14. H. S. Yoo, H. Myung, and H. J. Oh, "Issue Survival Cycle Analysis of Social Disasters in Social Media", Journal of KIIT, Vol. 15, No. 7, pp. 139-147, Jul. 2017.

Cited by

  1. Disaster Response Content for Effective Evacuation of Disaster Vulnerable Classes in Case of Disaster vol.19, pp.12, 2017, https://doi.org/10.9728/dcs.2018.19.12.2267