한국방재안전학회논문집 (Journal of Korean Society of Disaster and Security)

  • 제14권3호
  • /
  • Pages.17-27
  • /
  • 2021
  • /
  • 2466-1147(pISSN)
  • /
  • 2508-285X(eISSN)
한국방재안전학회 (Korean Society of Disaster & Security)
권호 표지
DOI QR코드

DOI QR Code

기계학습기법을 이용한 땅밀림 위험등급 분류

Classification of Soil Creep Hazard Class Using Machine Learning

  • 이기하 (경북대학교 미래과학기술융합학과) ;
  • 레수안히엔 (경북대학교 재난대응전략연구소) ;
  • 연민호 (경북대학교 미래과학기술융합학과) ;
  • 서준표 (국립산림과학원 산불.산사태연구과) ;
  • 이창우 (국립산림과학원 산불.산사태연구과)
  • Lee, Gi Ha (Dept. of Advanced Science and Technology Convergence, Kyungpook National University) ;
  • Le, Xuan-Hien (Disaster Prevention Emergency Management Institute, Kyungpook National University) ;
  • Yeon, Min Ho (Dept. of Advanced Science and Technology Convergence, Kyungpook National University) ;
  • Seo, Jun Pyo (Division of Forest Fire and Landslide, National Institute of Forest Science) ;
  • Lee, Chang Woo (Division of Forest Fire and Landslide, National Institute of Forest Science)
  • 투고 : 2021.08.15
  • 심사 : 2021.09.12
  • 발행 : 2021.09.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 연구에서는 6개의 기계학습 기법들을 활용하여 2019년과 2020년 전국 땅밀림 현장조사 결과를 기반으로 땅밀림 위험지역을 A부터 C까지 3개 등급(A등급: 위험, B등급: 보통, C등급: 양호)으로 구분할 수 있는 분류모형을 구축하고, 분류 정확도를 비교·분석한다. 기계학습 기법으로는 K-Nearest Neighbor, Support Vector Machine, Logistic Regression, Decision Tree, Random Forest, Extreme Gradient Boosting 총 6개를 적용하였다. 분류 정확도 분석결과, 6개의 기법 모두 0.9 이상의 우수한 정확도를 보여주었다. 수치형 자료를 학습에 적용한 경우가, 문자형 자료를 학습한 모형보다 우수한 성능을 나타냈으며, 현장조사 평가점수 자료군(C1~C4) 보다는 전문가의견이 반영된 평가점수 자료군(R1~R4)으로 학습한 모형이 정확도가 높은 것으로 분석되었다. 특히, 직접징후와 간접징후 정보를 학습에 반영한 경우가 예측정확도가 높게 나타났다. 향후 땅밀림 현장조사 자료가 지속적으로 확보될 경우, 본 연구에서 활용한 기계학습기법은 땅밀림 분류를 위한 도구로 활용이 가능할 것으로 판단된다.

In this study, classification models were built using machine learning techniques that can classify the soil creep risk into three classes from A to C (A: risk, B: moderate, C: good). A total of six machine learning techniques were used: K-Nearest Neighbor, Support Vector Machine, Logistic Regression, Decision Tree, Random Forest, and Extreme Gradient Boosting and then their classification accuracy was analyzed using the nationwide soil creep field survey data in 2019 and 2020. As a result of classification accuracy analysis, all six methods showed excellent accuracy of 0.9 or more. The methods where numerical data were applied for data training showed better performance than the methods based on character data of field survey evaluation table. Moreover, the methods learned with the data group (R1~R4) reflecting the expert opinion had higher accuracy than the field survey evaluation score data group (C1~C4). The machine learning can be used as a tool for prediction of soil creep if high-quality data are continuously secured and updated in the future.

키워드

과제정보

This subject is supported by Korea Ministry of Environment as "The SS projects; 2019002830001".

참고문헌

  1. Akgun, A. (2012). A Comparison of Landslide Susceptibility Maps Produced by Logistic Regression, Multi-Criteria Decision, and Likelihood Ratio Methods: A Case Study at Izmir, Turkey. Landslides. 9(1): 93-106. https://doi.org/10.1007/s10346-011-0283-7
  2. Althuwaynee, O. F., Pradhan, B., Park, H. J., and Lee, J. H. (2014). A Novel Ensemble Decision Tree-based CHi-squared Automatic Interaction Detection (CHAID) and Multivariate Logistic Regression Models in Landslide Susceptibility Mapping. Landslides. 11(6): 1063-1078. https://doi.org/10.1007/s10346-014-0466-0
  3. Bergen, K. J., Johnson, P. A., Maarten, V., and Beroza, G. C. (2019). Machine Learning for Data-driven Discovery in Solid Earth Geoscience. Science. 363(6433).
  4. Byeon, S. H., Kang, H. J., Han, J. W., and Kim, T. W. (2008). Flood Mitigation Planing for a Basin Using a Decision Tree Model. Journal of Civil and Environmental Engineering Research B. 28(1B): 33-40.
  5. Chae, B. G., Kim, W. Y., Kim, Y. C., Kim, K. S., Lee, C. O. and Choi, Y. S. (2004). Development of a Logistic Regression Model for Probabilistic Prediction of Debris Flow. The Journal of Engineering Geology. 14(2): 211-222.
  6. Chen, W., Peng, J., Hong, H., Shahabi, H., Pradhan, B., Liu, J., Zhu, A., Pei, X., and Duan, Z. (2018). Landslide Susceptibility Modelling using GIS-based Machine Learning Techniques for Chongren County, Jiangxi Province, China. Science of the total environment. 626: 1121-1135. https://doi.org/10.1016/j.scitotenv.2018.01.124
  7. Choi, S. W., Jang, W. C. (2017). Forecasting Probabilities of Earthquake in Korea Based on Seismological Data. The Korean Journal of Applied Statistics. 30(5): 759-774. https://doi.org/10.5351/KJAS.2017.30.5.759
  8. Danneels, G., Pirard, E., and Havenith, H. B. (2007). Automatic Landslide Detection from Remote Sensing Images using Supervised Classification Methods. In 2007 IEEE International Geoscience and Remote Sensing Symposium. 3014-3017.
  9. Ding, A., Zhang, Q., Zhou, X., and Dai, B. (2016). Automatic Recognition of Landslide Based on CNN and Texture Change Detection. In 2016 31st Youth Academic Annual Conference of Chinese Association of Automation (YAC). 444-448.
  10. Kirschbaum, D. and Stanley, T. (2018). Satellite-based Assessment of Rainfall-triggered Landslide Hazard for Situational Awareness. Earth's Future. 6(3): 505-523. https://doi.org/10.1002/2017EF000715
  11. Ma, Z., Mei, G., and Piccialli, F. (2020). Machine Learning for Landslides Prevention: A Survey. Neural Computing and Applications. 1-27.
  12. National Institute of Forest Science. (2018). Field Survey Manual of Soil Creep. Seoul: NIFoS.
  13. Park, J. H. (2015). Analysis on the Characteristics of the Landslide-with a Special Reference on Geo-topographical Characteristics. Journal of Korean Society of Forest Science. 104(4): 588-597. https://doi.org/10.14578/jkfs.2015.104.4.588
  14. Segoni, S., Lagomarsino, D., Fanti, R., Moretti, S., and Casagli, N. (2015). Integration of Rainfall Thresholds and Susceptibility Maps in the Emilia Romagna (Italy) Regional-scale Landslide Warning System. Landslides. 12(4): 773-785. https://doi.org/10.1007/s10346-014-0502-0
  15. Sit, M., Demiray, B. Z., Xiang, Z., Ewing, G. J., Sermet, Y., and Demir, I. (2020). A Comprehensive Review of DEEP Learning Applications in Hydrology and Water Resources. Water Science and Technology. 82(12): 2635-2670. https://doi.org/10.2166/wst.2020.369
  16. Woo, S. Y., Jung, C. G., Kim, J. U. and Kim, S. J. (2018). Assessment of Climate Change Impact on Aquatic Ecology Health Indices in Han River Basin using SWAT and Random Forest. Journal of Korea Water Resources Association. 51(10): 863-874. https://doi.org/10.3741/JKWRA.2018.51.10.863