한국측량학회지 (Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography)

  • 제39권6호
  • /
  • Pages.351-361
  • /
  • 2021
  • /
  • 1598-4850(pISSN)
  • /
  • 2288-260X(eISSN)
한국측량학회 (Korean Society of Surveying, Geodesy, Photogrammetry and Cartography)
권호 표지
DOI QR코드

DOI QR Code

머신러닝을 이용한 경기도 화재위험요인 예측분석

Predictive Analysis of Fire Risk Factors in Gyeonggi-do Using Machine Learning

  • 투고 : 2021.07.28
  • 심사 : 2021.12.14
  • 발행 : 2021.12.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

화재는 막대한 재산과 인명피해를 초래하고 있으며 크고 작은 화재가 지속해서 발생하고 있다. 따라서 본 연구는 화재 유형별로 화재에 영향을 미치는 각종 위험요인을 예측하고자 한다. 전국에서 화재 발생 건수가 가장 많은 경기도를 대상으로 화재발생위험요인 예측분석을 실시하였다. 또한, 머신러닝 방법인 SVM, RF, GBRT를 활용하여 각 모형의 정확성을 MAE,RMSE를 통해 적합도가 높은 모형을 제시하였으며 이를 토대로 경기도 화재발생요인 예측분석을 실시하였다. 머신러닝 방법 3가지를 비교분석한 결과 RF가 MAE 1.517, RMSE 1.820으로 나타났으며 MAE, RMSE 검증데이터 및 시험데이터의 경우 MAE값 0.024, RMSE값 0.12의 차이로 매우 유사하게 나타나 가장 우수한 예측력으로 나타났다. RF기법을 적용하여 분석한 결과 공통적으로 발화장소가 화재발생에 가장 큰 영향을 주는 위험요인으로 나타났다. 이러한 연구 결과는 화재발생에 영향을 주는 요인들의 위험순서를 파악하여 화재안전관리의 유용한 자료로 활용될 것으로 예상된다.

The seriousness of fire is rising because fire causes enormous damage to property and human life. Therefore, this study aims to predict various risk factors affecting fire by fire type. The predictive analysis of fire factors was carried out targeting Gyeonggi-do, which has the highest number of fires in the country. For the analysis, using machine learning methods SVM (Support Vector Machine), RF (Random Forest), GBRT (Gradient Boosted Regression Tree) the accuracy of each model was presented with a high fit model through MAE (Mean Absolute Error) and RMSE (Root Mean Squared Error), and based on this, predictive analysis of fire factors in Gyeonggi-do was conducted. In addition, using machine learning methods such as SVM (Support Vector Machine), RF (Random Forest), and GBRT (Gradient Boosted Regression Tree), the accuracy of each model was presented with a high-fit model through MAE and RMSE. Predictive analysis of occurrence factors was achieved. Based on this, as a result of comparative analysis of three machine learning methods, the RF method showed a MAE = 1.765 and RMSE = 1.876, as well as the MAE and RMSE verification and test data were very similar with a difference between MAE = 0.046 and RMSE = 0.04 showing the best predictive results. The results of this study are expected to be used as useful data for fire safety management allowing decision makers to identify the sequence of dangers related to the factors affecting the occurrence of fire.

키워드

과제정보

본 연구는 과학기술정보통신부 이공학개인기초연구지원사업 기본연구(2021R1F1A106422811)의 연구비 지원에 의해 수행되었습니다.

참고문헌

  1. Brieman, L. (2001), Random Forests, Machine learning, Vol. 45, No. 1, pp. 5-32. (in English) https://doi.org/10.1023/A:1010933404324
  2. Bae, S.W. (2019), Forecasting Property Prices Using the Machine Learning Methods: Model Comparisons, Ph.D. dissertation, Dankuk University, Seoul, Korea, 177p.
  3. David, Rohde., Jonathan, Corcoran. and Prem, Chhetri. (2010), Spatial Forecasting of Residential Urban Fires: A Bayesian Approach, Environment and Urban Systems, Vol. 34, pp. 58-69. (in English) https://doi.org/10.1016/j.compenvurbsys.2009.09.001
  4. Hong, S.Y., Cho, S.H., Kim, M.S. and Moon, I. (2019), Fire Prediction based on Weather and Fire Data using Artificial Neural Network, Fire Science and Engineering, Vol. 19, No. 7, pp. 275-281. (in Korean with English abstract)
  5. Hwang, W.U., Go, M.H. and Yang, B.S. (2004), Cavitation Condition Monitoring of Butterfly Valve Using Support Vector Machine, Transactions of the Korean society for noise and vibration engineering, Vol. 14, No. 2, pp. 119-127. (in Korean with English abstract) https://doi.org/10.5050/KSNVN.2004.14.2.119
  6. Hastie, C. and Searle, R. (2016), Socio-Economic and Demographic Predictors of Accidental Dwelling Fire Rates, Fire Safety Journal, Vol. 84, pp. 50-56. (in English) https://doi.org/10.1016/j.firesaf.2016.07.002
  7. Jang, H.S. (2018), Predictive evaluation of fecal E. coli occurrence using machine learning, Ph.D. dissertation, Kunsan National University, Jeonbuk, Korea, 121p.
  8. Kim, M.J. (2014), A Study on Fire Prediction Model Development Using Data Mining, Master's thesis, Kangwon University, Kangwon, Korea, 52p.
  9. Kim, S.J. and Ahn, H.C. (2016), Application of Random Forests to Corporate Credit Rating Prediction, Journal of Industrial Innovation Research, Vol. 32, No. 1, pp. 187-211. (in Korean with English abstract)
  10. Kim, H.R., Sin, J.W., Park, Y.J., Lee, H.P. and Moon, K.A. (2010), A Study on the Statistical Analysis of Fire Patterns in Gangwon Province, Fire Science and Engineering, Proceedings of the Korea Institute of Fire Science and Engineering Conference, 23 April, Kyungmin University, Korea, pp. 419-423.
  11. Lee, C.L. (2015), Estimating Single-family House Prices Using Non-parametric spatial Models and an Ensemble Learning Approach, Ph.D. dissertation, Seoul National University, Seoul, Korea, 165p.
  12. Lee, H.L. (2020), Real-time Prediction of Penetration Rate of Shield TBM based on Machine Learning, Ph.D. dissertation, Inha University, Inchoen, Korea, 132p.
  13. Lee, C.L. and Park, K.H. (2016), Application of machine learning models for estimating house price, Journal of the Korean Geographical Society, Vol. 51, No. 2, pp. 219-333. (in Korean with English abstract)
  14. Lee, H.P., Lee, S.C., Hwang, M.J., Park, Y.J., Moon, K.A. and Kim, H.B. (2010), A Study on the Analysis of Fire Patterns using the Decision Tree Analysis Method, Fire Science and Engineering, Proceedings of the Korea Institute of Fire Science and Engineering Conference, 30 October, Seoul, Korea, pp. 349-353.
  15. Lu, S., Liang, C., Song, W. and Zhang, H. (2013), Frequency-size Distribution and Time-scaling Property of High-casualty Fires in China: Analysis and Comparison, Safety Science, Vol 5, pp. 209-216. (in English)
  16. National Fire Agency. (2019), National Fire Data System, https://www.nfds.go.kr/, 28 June 2021.
  17. Park, K.H. (2018), Bagging, Random Forest and Ensemble SVM Comparison studies, Master thesis, Inha University, Inchoen, Korea, 25p.
  18. Rifan, Ardianto1. and Prem, Chhetri1. (2019), Modeling Spatial-Temporal Dynamics of Urban Residential Fire Risk Using a Markov Chain Technique, Int J Disaster Risk Sci, Vol. 10, pp. 57-73. (in English) https://doi.org/10.1007/s13753-018-0209-2
  19. Seo, J.D. (2016), Foreign Exchange Rate Forecasting Using the GARCH extended Random Forest Model, Journal of Industrial Economics and Business, Vol. 29, No. 5, pp. 1607-1628. (in Korean with English abstract)
  20. Seo, M.S. and Yoo, H.H. (2020) Significance Analysis of Facility Fires Though Spatial Econometrics Assessment, Journal of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography, Vol. 38, No. 3, pp. 281-293. (in Korean with English abstract) https://doi.org/10.7848/KSGPC.2020.38.3.281
  21. Vapnik, V. and Chervonenkis, A. (1964), A Note on One Class of Perceptrons, Automation and Remote Control, Vol. 25, No. 1, pp. 103-109. (in English)
  22. Vapnik, V. and Lerner, A. (1963), Pattern Recognition Using Generalized Portrait Method, Automation and Remote Control, Vol. 24, No. 6, pp. 774-780. (in English)