All Issue

2021 Vol.31, Issue 4 Preview Page

Research Article

Machine-Learning Evaluation of Factors Influencing Landslides

머신러닝기법을 이용한 산사태 발생인자의 영향도 분석

Seong-Yong Park1
Seong-Woo Moon2
Jaewan Choi3
Yong-Seok Seo4*
박 성용1
문 성우2
최 재완3
서 용석4*
1Master Student, Department of Earth and Environmental Sciences, Chungbuk National University
2Post-Doctor, Department of Earth and Environmental Sciences, Chungbuk National University
3Professor, Department of Civil Engineering, Chungbuk National University
4Professor, Department of Earth and Environmental Sciences, Chungbuk National University
1충북대학교 지구환경과학과 석사과정
2충북대학교 지구환경과학과 박사 후 연구원
3충북대학교 토목공학과 교수
4충북대학교 지구환경과학과 교수
*Corresponding Author
31 December 2021. pp. 701-708
PDF XML Citation
Abstract

본 연구에서는 산사태가 다수 발생한 충주 산척면 지역을 대상으로 야외지질조사 및 일련의 실내시험을 수행하여 데이터를 취득하고, 이후 인공신경망(Artificial neural network)과 로지스틱 회귀분석(Logistic regression)을 적용하여 각 인자가 산사태 발생에 미치는 영향도를 분석하였다. 야외지질조사 시 산사태 발생 유무에 따라 불교란시료를 채취하였으며, 동적 콘 관입시험기를 이용하여 토심을 측정하였다. 실내시험은 미국 표준시험법인 ASTM 규정에 따라 진행되었으며, 인자간 다중공선성을 해결하기 위해 VIF(Variation inflation factor)를 산정하였다. 다중공선성 분석을 통해 총 9개 인자(전단강도, 암종, 토심, 포화함수비, 비중, 투수계수, USCS, 사면 경사, 고도)가 분석에 적용되었다. 추후 도출되는 각 인자별 영향도를 직접적으로 비교하기 위해서 데이터는 최소값 0, 최대값 1이 되도록 최소-최대 정규화한 후 로지스틱 회귀분석 및 인공신경망 분석에 적용되었다. 로지스틱 회귀분석 결과, 토심, 경사, 포화함수비, 전단강도 순으로 산사태 발생에 영향력이 크게 나타났으며, 인공신경망 분석 결과, 경사, 토심, 포화함수비, 전단강도 순으로 영향력이 크게 나타났다. 각 분석기법으로 산정된 영향도를 산술평균한 결과, 토심, 경사, 포화함수비, 전단강도가 상위 4개 인자로 선정되었으며, 이들의 영향도 합계는 약 70%로 분석되었다.

키워드
산사태 발생 영향인자
영향도 분석
로지스틱 회귀분석
인공신경망 분석

Geological field surveys and a series of laboratory tests were conducted to obtain data related to landslides in Sancheok-myeon, Chungju-si, Chungcheongbuk-do, South Korea where many landslides occurred in the summer of 2020. The magnitudes of various factors’ influence on landslide occurrence were evaluated using logistic regression analysis and an artificial neural network. Undisturbed specimens were sampled according to landslide occurrence, and dynamic cone penetration testing measured the depth of the soil layer during geological field surveys. Laboratory tests were performed following the standards of ASTM International. To solve the problem of multicollinearity, the variation inflation factor was calculated for all factors related to landslides, and then nine factors (shear strength, lithology, saturated water content, specific gravity, hydraulic conductivity, USCS, slope angle, and elevation) were determined as influential factors for consideration by machine learning techniques. Minimum-maximum normalization compared factors directly with each other. Logistic regression analysis identified soil depth, slope angle, saturated water content, and shear strength as having the greatest influence (in that order) on the occurrence of landslides. Artificial neural network analysis ranked factors by greatest influence in the order of slope angle, soil depth, saturated water content, and shear strength. Arithmetically averaging the effectiveness of both analyses found slope angle, soil depth, saturated water content, and shear strength as the top four factors. The sum of their effectiveness was ~70%.

Keywords
influential factor of landslide
evaluation of effectiveness
logistic regression analysis
artificial neural network analysis
References
1
ASTM D2216-10, 2010, Standard test methods for laboratory determination of water (moisture) content of soil and rock by mass, ASTM International, West Conshohocken, PA, DOI:10.1520/D2216-10. 10.1520/D2216-10
2
ASTM D3080-98, 1998, Standard test method for direct shear test of soils under consolidated drained conditions, ASTM International, West Conshohocken, PA, DOI: 10.1520/D3080-98. 10.1520/D3080-98
3
ASTM D422-63, 2007, Standard test method for particle-size analysis of soils, ASTM International, West Conshohocken, PA, DOI: 10.1520/D0422-63R07E02. 10.1520/D0422-63R07E02
4
ASTM D4318-00, 2000, Standard test methods for liquid limit, plastic limit, and plasticity index of soils, ASTM InterWest Conshohocken, PA, DOI: 10.1520/D4318-00. 10.1520/D4318-00
5
ASTM D854-10, 2010, Standard test methods for specific gravity of soil solids by water pycnometer, ASTM International, West Conshohocken, PA, DOI: 10.1520/D0854-10. 10.1520/D0854-10
6
Basheer, I.A., Hajmeer, M., 2000, Artificial neural networks: Fundamentals, computing, design, and application, Journal of Microbiological Methods, 43(3), 3-31. 10.1016/S0167-7012(00)00201-3
7
Chae, B.G., Kim, W.Y., Cho, Y.C., Kim, K.S., Lee, C.O., Choi, Y.S., 2004, Development of a logistic regression model for procabilistic prediction of debris flow, The Journal of Engineering Geology, 14(2), 211-222 (in Korean with English abstract).
8
Chae, B.G., Seo, T.S., 2010, Suggestion of an evaluation chart for landslide susceptibility using a quantification analysis based on canonical correlation, Economic and Environmental Geology, 43(4), 381-391 (in Korean with English abstract).
9
Choi, J.S., 2000, Modern statistical analysis using SPSS Ver 10, Bogdoo Publish, 514p (in Korean).
10
Choi, J.W., Lee, S., Min, G.D., Woo, I., 2004, Landslide susceptibility mapping and verification using the GIS and bayesian probability model in Boun, Economic and Environmental Geology, 37(2), 207-223 (in Korean with English abstract).
11
Choi, S.W., Park, H.N., Jung, S.G., 1997, A study on landslide using geographic information system in the western part of Puyeo area, Research Institute for Basic Science Kongju National University, 6, 69-81 (in Korean with English abstract).
12
Chung, W.G., Lee, S.Y., 2007, A study on the forecasting of the apartment price index using artificial neural networks, Housing Studies, 15(3), 39-64 (in Korean with English abstract).
13
Coulomb, C.A., 1776, Essai sur une application des regles de maximis et minimis quelques problemes de statique, relatits a la architecture, Memoires de Mathematique de la Academie Royale de Science, 7, 343-387.
14
Garson, G.D., 1991, Interpreting neural-network connection weights, AI Expert, 6(4), 46-51.
15
Hazen, A., 1892, Experiments upon the purification of sewage and water at the Lawrence experiment station, 24th Annual Report, Massachusetts State Board of Health, 910p.
16
Jang, D.H., Park, N.W., Chi, K.H., Kim, M.K., Chung, C.J., 2004, Landslide susceptibility analysis in the Boeun area using a GIS-based bayesian prediction model, Journal of the Korean Geomorphological Association, 11(3), 13-23 (in Korean with English abstract).
17
Jeong, K.S., 2018, Application of artificial neural network model to an analysis of the factors affecting the intention of the vulnerable class to move to Hangbok Housing in Incheon, Housing Studies, 26(3), 55-78 (in Korean with English abstract). 10.24957/hsr.2018.26.3.55
18
Jeong, M., Choi, H., Choi, J., 2020, Analysis of change detection results by UNet++ models according to the characteristics of loss function, Korean Journal of Remote Sensing, 36(5-2), 929-937 (in Korean with English abstract).
19
Jo, Y.C., Chea, B.G., Kim, W.Y., Chang, T.W., 2007, A modified logistic regression model for probabilistic prediction of debris flow at the granitic rock area and its application; landslide prediction map of Gangreung area, Economic and Environmental Geology, 40(1), 115-128 (in Korean with English abstract).
20
Jo, Y.J., 2018, Big data SPSS latest analysis techniques, Hannarae Publishing Company, 168p (in Korean).
21
Joo, G., Park, C., Im, H., 2020, Performance evaluation of machine learning optimizers, Journal of Institute of Korean Electrical and Electronics Engineers, 24(3), 766-776 (in Korean with English abstract).
22
Kang, W.P., Murai, H., Omura, H., Ma, H.S., 1986, On the determination of slope stability to landslide by quantification (II), Journal of Korean Society of Forest Science, 75(1), 32-37 (in Korean with English abstract).
23
Kang, W.P., Woo, B.M., 1985, Studies on the landslide disasters occurred in Munhyon-dong on July 5, 1985, Journal of Korean Society of Forest Science, 70(1), 77-83 (in Korean with English abstract).
24
Kim, J.Y., Park, H.J., 2013, A comparative study of Fuzzy relationship and ANN for landslide susceptibility in Pohang area, Economic and Environmental Geology, 46(4), 301-312 (in Korean with English abstract). 10.9719/EEG.2013.46.4.301
25
Kim, M.K., Kim, S.P., Nho, H.J., Sohn, H.G., 2017, Landslide susceptibility mapping by comparing GIS-based spatial models in the Java, Indonesia, Surveying and Geo-Spatial Information Engineering, 37(5), 927-940 (in Korean with English abstract).
26
Kim, S.E., Seo, I.W., 2017, A study on the improvement of generalization accuracy of artificial neural network modeling using factor analysis and cluster analysis: Application to flow and water quality prediction, Water for Future, 50(6), 31-40 (in Korean).
27
Kim, Y.J., Kim, W.Y., Yu, I.H., 1992, GIS technolgy for analysing regional geologic hazards (Landslides), The Journal of Engineering Geology, 2(2), 131-140.
28
Kim, Y.J., Kim, W.Y., Yu, I.H., 1993, Analysis of regional geologic hazards using GIS, The Journal of Korean Society for Geospatial Information Science, 1(1), 89-94 (in Korean with English abstract).
29
Kim, Y.J., Kim, W.Y., Yu, I.H., Park, S.H., Baek, J.H., Lee, H.W., 1991, Analysis of regional geologic hazards using Geographic Information System, Korean Journal of Remote Sensing, 7(2), 165-178 (in Korean with English abstract).
30
Kim, Y.J., Yu, I.H., Kim, W.Y., Lee, S., Sin, E.S., Song, M.Y., 1996, A GIS technique to evaluate landslide activity, The Journal of Geographic Information System Association of Korea, 4(1), 83-92 (in Korean with English abstract).
31
KMA (Korea Meteorological Administration), 2020, KMA weather data service open MET data portal, Retrieved from https://data.kma.go.kr/data/grnd/selectAsosRltmList.do?pgmNo=36&tabNo=1.
32
Koo, H.B., Koo, J.D., 1995, Landslide data base system using GIS technology, The Journal of Geographic Information System Association of Korea, 3(1), 89-90 (in Korean with English abstract).
33
Kutner, M.H., Nachtsheim, C.J., Neter, J., 2004, Applied linear regression models, 4th ed., McGraw-Hill Education, 701p.
34
Lee, I.M., 1987, A stochastic numerical analysis of groundwater fluctuations in hillside slopes for assessing risk of landslides, Journal of the Korean Geotechnical Society, 3(4), 41-54.
35
Lee, M.J., Lee, S.R., Jeon, S.W., 2012, Landslide hazard mapping and verfication using probability rainfall and artificial neural networks, Journal of the Association Geographic Information Studies, 15(2), 57-70 (in Korean with English abstract). 10.11108/kagis.2012.15.2.057
36
Lee, S., Lee, M.J., Won, J.S., 2005, Landslide susceptibility analysis and verification using artificial neural network in the Kangneung area, Economic and Environmental Geology, 38(1), 33-43 (in Korean with English abstract).
37
Lee, S., Oh, H.J., 2019, Landslide susceptibility prediction using evidential belief function, weight of evidence and artificial neural network models, Korean Journal of Remote Sensing, 35(2), 229-316 (in Korean with English abstract).
38
Lee, S., Ryu, J.H., Min, K.D., Won, J.S., 2000, Landslide susceptibility analysis using artificial neural networks, Economic and Environmental Geology,33(4), 333-340 (in Korean with English abstract).
39
Lee, S.W., 1979, Studies on the causal factors of landslides on limestone soils in Pyeongchang-kun, Research Reports of Agricultural Science and Technology, 6(2), 125-133 (in Korean with English abstract).
40
Lee, W.Y., 2018, Determination method of hyperparameters based on HS algorithm for design of optimal convolutional neural network, Chung-Ang University, 1-3 (in Korean with English abstract).
41
Lee, Y.J., Park, G.A., Kim, S.J., 2006, Analysis of landslide hazard area using logistic regression analysis and AHP (Analytical hierarchy process) approach, The Journal of the Korean Society of Civil Engineers, 26(5D), 861-867 (in Korean with English abstract).
42
Li, L., Kim, H.L., Jun, K.S., Choi, M.H., 2016, Estimation of river discharge using satellite-derived flow signals and artificial neural network model: Application to Imjin river, The Journal of Korea Water Resources Association, 49(7), 589-597 (in Korean with English abstract). 10.3741/JKWRA.2016.49.7.589
43
Ma, S.K., 1979, Environmental interpretation on soil mass movement spot and disaster dangerous site for precautionary measures -in Peong Chang Area-, Journal of Korean Society of Forest Science, 45(1), 11-25 (in Korean with English abstract).
44
Mckelvey, R.D., Zavoina, W., 1975, A statistical model for the analysis of ordinal level dependent variables, Journal of Mathematical Sociology, 4(1), 103-120. 10.1080/0022250X.1975.9989847
45
Muthén, B.O., 1984, A general structural equation model with dichotomous, ordered categorical, and continuous latent variable indicators, Psychometrika, 49(1), 115-132. 10.1007/BF02294210
46
NGII (National Geographic Information Institute), 2019, Digital topographic map, Retrieved from http://map.ngii.go.kr/ms/map/NlipMap.do.
47
Oh, H.J., 2010, Landslide detection and landslide susceptibility mapping using aerial photos and artificial neural networks, Korean Journal of Remote Sensing, 26(1), 47-57 (in Korean with English abstract).
48
Park, B.S., Yeo, S.C., 1971, Explanatory text of the geological map of Korea: Moggye sheet, Geological Survey of Korea, 24p.
49
Park, S.J., 2014, Generality and specificity of landforms of the Korean peninsula, and its sustainability, Journal of the Korean Geographical Society, 49(5), 656-674 (in Korean with English abstract).
50
Park, Y.G., 1965, Studies on the land slide, The Journal of The Korean Society of Agricultural Engineers, 7(1), 43-48 (in Korean with English abstract).
51
Quan, H.C., Lee, B.G., Cho, E.I., 2008, Landslide susceptibility analysis in Jeju using artificial neural network (ANN) and GIS, Journal of the Environmental Sciences, 17(6), 679-687 (in Korean with English abstract). 10.5322/JES.2008.17.6.679
52
Quan, H.C., Lee, B.G., Lee, C.S., Ko, J.W., 2011, The landslide probability analysis using logistic regression analysis and artificial neural network methods in Jeju, Journal of the Korean Society for Geospatial Information Science, 19(3), 33-40 (in Korean with English abstract).
53
Rao, V.B., Rao, H.V., 1993, C++ neural network and fuzzy logic, Management Information, 408p.
54
Ro, K.S., Jeon, B.J., Jeon, K.W., 2015, Induction wall influence review by debris flow's impact force, Journal of the Korean Society of Hazard Mitigation, 15(2), 159-164 (in Korean with English abstract). 10.9798/KOSHAM.2015.15.2.159
55
Ryu, J.H., Lee, S., Won, J.S., 2002, Weight determination of landslide factors using artificial neural networks, Economic and Environmental Geology 35(1), 67-74 (in Korean with English abstract).
56
Ryu, S.G., 2008, Effects of multicollinearity in logit model, Journal of Korean Society of Transportation, 26(1), 113-126 (in Korean with English abstract).
57
Sidle, R.C., Ziegler, A.D., Negishi, J.N., Nik, A.R., Siew, R., Turkelboom, F., 2006, Erosion processes in steep terrain-Truths, myths, and uncertainties related to forest management in Southeast Asia, Forest Ecology and Management, 224, 199-225. 10.1016/j.foreco.2005.12.019
58
Šimundić, A.M., 2009, Measures of diagnostic accuracy: Basic definitions, The Journal of the International Federation of Clinical Chemistry and Laboratory Medicine, 19(4), 203-211.
59
Song, C.H., Lee, J.S., Kim, Y.T., 2021, Analysis of landslide susceptibility using deep neural network, Journal of the Korean Society of Hazard Mitigation, 21(3), 141-150 (in Korean with English abstract). 10.9798/KOSHAM.2021.21.3.141
60
Song, Y.S., Jo, Y.C., 2013, A study on the possibility of landslide damage in the construction section of railway tunnel, Journal of the Korean Geosynthetics Society, 12(2), 17-24 (in Korean).
61
Urmi, R.B.R., Al, M., Jang, D.H., 2020, Life risk assessment of landslide disaster in Jinbu area using logistic regression model, Journal of the Korean Geomorphological Association, 27(2), 65-80. 10.16968/JKGA.27.2.65
62
Woo, B.M., 1984, Landslide disaster countermeasures in Korea, Journal of Korean Society of Forest Science, 63(1), 51-60 (in Korean with English abstract).
63
Woo, C.S., Kwon, H.J., Lee, C.W., Kim, K.H., 2014, Landslide hazard prediction map based on logistic regression model for applying in the whole country of South Korea, Journal of the Korean Society of Hazard Mitigation, 14(6), 117-123 (in Korean with English abstract). 10.9798/KOSHAM.2014.14.6.117
64
Yang, A., Park, H.D., 2021, Comparison of landslide susceptibility assessments using logistic regression with a study site extent scale, Journal of The Korean Society of Mineral and Energy Resources Engineers, 58(2), 87-99 (in Korean with English abstract). 10.32390/ksmer.2021.58.2.087
65
Yang, I.T., Chun, K.S., Lee, S.Y., 2005, The application of GIS and AHP for landslide vulnerable estimation, Journal of Industrial Technology, Kandgwon National University, 25, 47-54 (in Korean with English abstract).
Information
  • Publisher :Korean Society of Engineering Geology
  • Publisher(Ko) :대한지질공학회
  • Journal Title :The Journal of Engineering Geology
  • Journal Title(Ko) :지질공학
  • Volume : 31
  • No :4
  • Pages :701-708
  • Received Date :2021. 12. 15
  • Revised Date :2021. 12. 21
  • Accepted Date : 2021. 12. 21
  • DOI :https://doi.org/10.9720/kseg.2021.4.701
Journal Informaiton The Journal of Engineering Geology The Journal of Engineering Geology
Online Submission
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close