재해기상 언론기사 빅데이터를 활용한 피해정보 자동 분류기 개발
Developing and Evaluating Damage Information Classifier of High Impact Weather by Using News Big Data
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Recently, the importance of impact-based forecasting has increased along with the socio-economic impact of severe weather have emerged. As news articles contain unconstructed information closely related to the people’s life, this study developed and evaluated a binary classification algorithm about snowfall damage information by using media articles text mining. We collected news articles during 2009 to 2021 which containing ‘heavy snow’ in its body context and labelled whether each article correspond to specific damage fields such as car accident. To develop a classifier, we proposed a probability-based classifier based on the ratio of the two conditional probabilities, which is defined as I/O Ratio in this study. During the construction process, we also adopted the n-gram approach to consider contextual meaning of each keyword. The accuracy of the classifier was 75%, supporting the possibility of application of news big data to the impact-based forecasting. We expect the performance of the classifier will be improve in the further research as the various training data is accumulated. The result of this study can be readily expanded by applying the same methodology to other disasters in the future. Furthermore, the result of this study can reduce social and economic damage of high impact weather by supporting the establishment of an integrated meteorological decision support system.
Abstract
Recently, the importance of impact-based forecasting has increased along with the socio-economic impact of severe weather have emerged. As news articles contain unconstructed information closely related to the people’s life, this study developed and evaluated a binary classification algorithm about snowfall damage information by using media articles text mining. We collected news articles during 2009 to 2021 which containing ‘heavy snow’ in its body context and labelled whether each article correspond to specific damage fields such as car accident. To develop a classifier, we proposed a probability-based classifier based on the ratio of the two conditional probabilities, which is defined as I/O Ratio in this study. During the construction process, we also adopted the n-gram approach to consider contextual meaning of each keyword. The accuracy of the classifier was 75%, supporting the possibility of application of news big data to the impact-based forecasting. We expect the performance of the classifier will be improve in the further research as the various training data is accumulated. The result of this study can be readily expanded by applying the same methodology to other disasters in the future. Furthermore, the result of this study can reduce social and economic damage of high impact weather by supporting the establishment of an integrated meteorological decision support system.
참고문헌 (31)
- Choe, H. S., "How to utilize big data in the public sector", It’s Smart Media, 2(3), pp.18~25 (2013)
- Choi, S. H., "A new solution to disaster management, Big data", Meteorological Technology & Policy, 6(2), pp.77~87 (2013)
- Chung, K. Y., "Visions and directions for the Impact-based Forecasting", Meteorological Technology & Policy, 9(1), pp.6~22 (2016)
- Cox, D.R., "The regression analysis of binary sequences", Journal of the Royal Statistical Society: Series B (Methodological), 20(2), pp.215~232 (1958)
- Gupta V., "A Survey of Text Mining Techniques and Applications", Journal of Emerging Technologies in Web Intelligence, 1, pp.60~76 (2009)
- 한승룡, "복합위험요소에 대한 도시지역의 재해위험도 평가", 한국방재학회논문집, 15(3), pp.33~43 (2015), 한국방재학회
- Ihm, H., "Multi-source food hazard event extraction for public health", Proceedings of 2017 IEEE International Conference on Big Data and Smart Computing (BigComp), pp.414~417 (2017)
- 정재인, "뉴스 기사 텍스트 마이닝과 네트워크 분석을 통한 폭염의 사회·경제적 영향 유형 도출: 2012~2016년 사례", 대기, 30(3), pp.237~248 (2020), 한국기상학회[DOI]
- 강은숙, "직무분석을 위한 머신러닝 기반 전자문서 자동 분류 모델", 한국정보기술학회논문지, 17(7), pp.23~29 (2019), 한국정보기술학회[DOI]
- 김연우, "실시간 기상 빅데이터를 활용한 홍수 재난안전 시스템 설계 및 구현", 한국콘텐츠학회 논문지, 17(1), pp.351~362 (2017), 한국콘텐츠학회[DOI]
- Kohavi, R., "A study of cross-validation and bootstrap for accuracy estimation and model selection", International Joint Conference on Artificial Intelligence, 14(2), pp.1137~1145 (1995)
- Building and evaluating climate change adaptation capacity for national risk man-agement: Analysis of direct and indirect effects of heat waves and cold waves based on data, "Korea Environment Institute", ~ (2019)
- Planning research on integrated data construction methods for forecasting the impact of meteorological disasters, "Korea Meteorological Administration", ~ (2016)
- 권세혁, "머신러닝 기법을 활용한 대용량 시계열 데이터 이상 시점탐지 방법론 : 발전기 부품신호 사례 중심", 한국산업경영시스템학회지, 43(2), pp.33~38 (2020), 한국산업경영시스템학회[DOI]
- 권수정, "119 신고 데이터를 이용한 자연어처리 기반 재난안전 상황 분류 알고리즘 분석", 정보처리학회 논문지, 9(10), pp.317~322 (2020), 한국정보처리학회[DOI]
- Lee, K. K., "Measures to improve reliabiligy of weather forecasts based on big data analysis", Meteorological Technology & Policy, 6(2), pp.32~46 (2013)
- 이기광, "Twitter를 활용한 기상예보서비스에 대한 사용자들의 만족도 분석", 한국산업경영시스템학회지, 41(2), pp.9~15 (2018), 한국산업경영시스템학회[DOI]
- 이기광, "A Business Application of the Business Intelligence and the Big Data Analytics", 한국산업경영시스템학회지, 42(4), pp.84~90 (2019), 한국산업경영시스템학회[DOI]
- 이기광, "언론기사 빅데이터를 활용한 대설 재해기상의 영향예보 위험수준 산정 방안 연구", e-비즈니스연구, 23(1), pp.233~245 (2022), 국제e-비즈니스학회
- Unstructured big data disaster safety information pattern analysis, "National Disaster Management Research Institute", ~ (2017)
- Research on the use of disaster weather information using Big data, "National Institute of Meteorological Sciences", ~ (2019)
- Noh, B. J., "Trend analysis of foot-and-mouth disease using keyword network", Proceedings of Conference on Korean Society for Internet Information, 17(1), pp.217~218 (2016)
- 오효정, "소셜 빅데이터 기반 사회적 이슈 리스크 유형 분류", 한국콘텐츠학회 논문지, 16(8), pp.1~9 (2016), 한국콘텐츠학회[DOI]
- Palmer, T. N., "The economic value of ensemble forecasts as a tool for risk assessment: From days to decades", Quarterly Journal of the Royal Meteorological Society:A Journal of the Atmospheric Sciences, Applied Meteorology and Physical Oceanography, 128(581), pp.747~774 (2002)
- 박종철, "폭염에 의한 축산폐사와 뉴스 빅데이터의 상관관계 분석", 한국지리학회지, 8(3), pp.529~543 (2019), 한국지리학회[DOI]
- Seo, J. H., "Assessing likelihood of drought impact occurrence in South Korea through Machine Learning", Proceedings of the Korea Water Resources Association Conference, pp.77~77 (2021)
- 신동희, "국내 재난관리 분야의 빅 데이터 활용 정책방안", 한국콘텐츠학회 논문지, 15(2), pp.377~392 (2015), 한국콘텐츠학회[DOI]
- "80 Percent of Your Data Will Be Unstructured in Five Years, Data Management Solutions Review" (2019)[URL]
- Vapnik, V., "A class of algo-rithms for pattern recognition learning", Avtomat. i Telemekh, 25(6), pp.937~945 (1964)
- Post-Typhoon Haiyan (Yolanda) Expert Mission to the Philippines, Manila and Tacloban, "WMO", ~ (2014)
- Yeh, S. W., "Suggestions for expanding impact forecasting services", Meteorological Technology & Policy, 10(1), pp.6~17 (2017)