THE APPLICABILITY OF RANDOM FOREST REGRESSION METHOD FOR THE PREDICTION OF THE CONSISTENCY AND COMPACTION PROPERTIES OF SOILS

Abstract

Within this paper, the relationship between consistency limits and compaction characteristics of highly plastic clay soils was examined by comparative analysis of regression and Random Forest regression methods. Due to the difficulty of standard laboratory experiments that have long sample preparation-waiting processes, which are used to directly obtain the maximum dry unit weight and optimum water content values representing the compaction parameters of soils, it is relatively more applicable to estimate these parameters by the use of practical experiments is a method that is frequently applied today. In addition, the liquid limit, one of the consistency limit tests, is an experiment that is applied in all geotechnical engineering designs and gives satisfactory results. In this study, a two-stage estimation process was carried out by using a database created by using 387 consistency limit and 59 compaction-consistency limit test couples presented in the literature studies of high plasticity clay soils. In the first stage, the estimation of the plasticity index directly from the liquid limit, in the second stage, the probability of estimating the compaction parameters from the plasticity index was investigated. At the same time, this study is focused on the inconsistency of the real data obtained directly from the laboratory experiments and the low accuracy rate that occurs in the general regression studies due to the fact that these data do not follow a certain trend. It is also examined how these accuracy rates can be increased by the Random Forest regression method. Consequently, it is shown that the Random Forest regression method can be used for the estimation of the consistency and compaction properties of highly plastic clayey soils, and gives satisfactory results to use.

Keywords

• Liquid Limit
• Compaction
• Clay Soils
• Random Forest Regression
• Decision Trees
• Regression method

Project Number

BYP-2020-34856 ve FBA-2020-34051

ZEMİNLERİN KIVAM VE KOMPAKSİYON ÖZELLİKLERİNİN TAHMİNİNDE RASTGELE ORMAN REGRESYONU YÖNTEMİNİN UYGULANABİLİRLİĞİ

Abstract

Bu makalede, yüksek plastisiteli kil zeminlerin kıvam limitleri ve kompaksiyon karakteristikleri arasındaki ilişki basit regresyon ve karar ağaçları tabanlı Rastgele Orman regresyon (RO) yöntemlerinin karşılaştırmalı olarak analiz edilmesi yoluyla irdelenmiştir. Zeminlerin kompaksiyon parametrelerini oluşturan maksimum kuru birim hacim ağırlık ve optimum su muhtevası değerlerinin doğrudan belirlenmesinde kullanılan standart laboratuvar deneylerin zorluğu ve uzun numune hazırlama-bekleme süreçleri içermesi nedeni ile göreceli olarak daha pratik deneyler kullanılarak bu parametrelerin tahmin edilmesi günümüzde sıklıkla uygulanılan bir yöntemdir. Ayrıca, kıvam limiti deneylerinden likit limit, tüm geoteknik mühendisliği tasarımlarında uygulanan ve tatminkar sonuçlar veren bir deneydir. Bu çalışmada, yüksek plastisiteli kil zeminlere ait literatürde sunulan 387 adet kıvam limiti ve 59 kompaksiyon-kıvam limiti test çiftinin kullanılması ile oluşturulan bir veri tabanı kullanılarak iki aşamalı bir tahmin süreci yürütülmüştür. Birinci aşamada plastisite indisinin doğrudan likit limit değerinden, ikinci aşamada ise kompaksiyon parametrelerinin plastisite indisinden tahmin olasılığı araştırılmıştır. Aynı zamanda, laboratuvar deneylerinden elde edilen gerçek verilerin tutarsızlık durumları ve bu verilerin belirli bir eğilim izlememesi sebebi ile genel regresyon çalışmalarında oluşan doğruluk oranı düşüklüğüne dikkat çekilerek, bu doğruluk oranlarının Rastgele Orman regresyonu yöntemi ile nasıl yükseltilebileceği de incelenmektedir. Sonuçlarda, Rastgele Orman regresyonu yönteminin yüksek plastisiteli kil zeminlerin kıvam ve kompaksiyon özelliklerinin tahmininde başarılı olduğu ve kullanılabilir nitelikte sonuçlar sunduğu gösterilmektedir.

Keywords

• Likit limit
• Kompaksiyon
• Kil Zeminler
• Rastgele Orman Regresyonu
• Karar Ağaçları
• Regresyon Yöntemi

Supporting Institution

İstanbul Üniversitesi-Cerrahpaşa

Project Number

BYP-2020-34856 ve FBA-2020-34051

Thanks

Bu çalışma, İstanbul Üniversitesi-Cerrahpaşa Bilimsel Araştırma Projeleri Birimi tarafından BYP-2020-34856 ve FBA-2020-34051 projeleri ile desteklenmektedir.

References

1. Akar, Ö., Güngör, O., 2012. Rastgele Orman Algoritması Kullanılarak Çok Bantlı Görüntülerin Sınıflandırılması. Jeodezi ve Jeoinformasyon Dergisi. 1 (2), 139-146. DOI: 10.9733/jgg.241212.1t.
2. Akar, Ö., Güngör, O., Akar, A., 2010. Rastgele Orman Sınıflandırıcısı ile Arazi Kullanım Alanlarının Belirlenmesi. III. Uzaktan Algılama ve Coğrafi Bilgi Sistemleri Sempozyumu, Gebze, Kocaeli, 142-152.
3. Akbay Arama, Z., Akin, M.S., Nuray, S.E., Dalyan, İ., 2020. Estimation of Consistency Limits of Fine-Grained Soils via Regression Analysis: A Special Case for High and Very High Plastic Clayey Soils in Istanbul. International Advanced Researchers and Engineering Journal (10.09.2020-Kabul edilmiş makale).
4. Akman, M., Genç, Y., Ankaralı, H., 2011. Random Forests Yöntemi ve Sağlık Alanında Bir Uygulama. Türkiye Klinikleri Biyoistatistik Dergisi. 3(1) 36–48.
5. Archer, K.J., 2008. Emprical Characterization of Random Forest Variable Importance Measure, Computational Statistical Data Analysis, Computational Statistics & Data Analysis, 52(4), 2249-2260.
6. Ardakani, A., Kordnaeij, A., 2019. Soil Compaction Parameters Prediction Using GMDH-Type Neural Network and Genetic Algorithm. European Journal of Environmental and Civil Engineering 23:4, 449-462.
7. ASTM Standard D 4318. Standard Test Methods for Liquid Limit, Plastic Limit and Plasticity Index of Soils.
8. ASTM Standard D 698-12. Standard Test Methods for Laboratory Compaction Characteristics of Soil Using Standard Effort.

9. Benson, C.H., Zhai, H., Wang, X., 1994. Estimating Hydraulic Conductivity of Compacted Clay Liners. Journal of Geotechnical Engineering Vol. 120, No. 2.
10. Benson, C.H., Daniel, D.E., Boutwell, G.P., 1999. Field Performance of Compacted Clay Liners. Journal of Geotechnical and Geoenvironmental Engineering Vol. 125, No.5.
11. Breiman, L., 2001. Random Forests, Machine Learning, 2001 Kluwer Academic Publishers, 45(1), 5-32.
12. Breiman, L., Cutler, A., 2005. Random Forest, http://www.stat.berkeley.edu/~g/RandomForests/cc_home.htm.
13. Boulila, W., Farah, I.R., Ettabaa, K.S., Solaiman, B., Ben Ghezala, H., 2011. A Data Mining Based Approach To Predict Spatiotemporal Changes in Satellite Images. International Journal of Applied Earth Observation and Geoinformation, 13, 386-395.
14. Canillas, C.E., Saloke M.V., 2001. Regression Analysis of Some Factors Influencing Soil Compaction. Soil & Tillage Research 61, 167–178.
15. Dewoolkar, M.M., Huzjak, R.J., 2005. Drained Residual Shear Strength of Some Claystones from Front Range, Colorado.
16. Dharumarajan, S., Hegdea, R., Singh, S.K., 2017. Spatial Prediction of Major Soil Properties Using Random Forest Techniques - A Case Study In Semi-Arid Tropics Of South India. Geoderma Regional 10 (2017) 154–162.
17. Fidan, H., 2020. Random Forest (Rastgele Orman) Algoritması Temelli Süreç İzleme Yönteminin Ambulatuar Kan Basıncı İzlemede Hipertansiyonun Erken Tanısı İçin Kullanımı. Yüksek Lisans Tezi, Muğla Sıtkı Kocaman Üniversitesi.
18. Grimm, R., Behrens, T., Märker, M., Elsenbeer, H., 2008. Soil Organic Carbon Concentrations and Stocks on Barro Colorado Island-Digital Soil Mapping Using Random Forests Analysis. Geoderma 146, 102–113.
19. Gislason, P.O., Benediktsson, J.A, Sveinsson, J.R., 2004. Random Forest Classification of Multi-Source Remote Sensing and Geographic Data. IEEE International Geoscienceand Remote Sensing Symposium IGARSS ‘04Proceedings, 2, 1049 – 1052.
20. Gislason, P.O., Benediktsson, J.A., Sveinsson, J.R., 2006. Random Forest For Land Cover Classification. Pattern Recognition Letters, 27, 294-300.
21. Gunaydın, O., 2009. Estimation of Soil Compaction Parameters by Using Statistical Analyses and Artifical Neural Networks. Environment Geology 57:203–215.
22. Gunther, E.C., Gerwien, R.W., Heyes, M.P., Bento, P., Stone, D.J. 2003. Prediction of Clinical Drug Efficacy by Classification of Drug-Induced Genomic Expression Profiles in Vitro. Proceedings of The National Academy of Sciences. 100(16) 9608-9613. https://doi.org/10.1073/pnas.1632587100.
23. Hill, T., Lewicki, P., 2006. Statistics Methods and Applications. A Comprehensive Reference For Science, Industry and Data Mining, StatSoft, Tulsa, OK.
24. Kavzaoglu, T., Sahin, E.S., Colkesen, I., 2012. Heyelan Duyarlılığının İncelenmesinde Regresyon Ağaçlarının Kullanımı: Trabzon Örneği (Assessment of Landslide Susceptibility Using Regression Trees: The Case of Trabzon Province)
25. Kemppinen, J., Niittynen, P., Riihimäki, H., Luoto, M., 2018. Modelling Soil Moisture In A High-Altitude Landscape Using LiDAR and Soil Data. Earth Surface Processes And Landforms 43, 1019–1031. DOI: 10.1002/esp.4301.
26. Khuntia, S., Mujtaba, H., Patra, C., Farooq, K., Sivakugan, N.&Das, B.M., 2014. Prediction of Compaction Parameters of Coarse Grained Soil Using Multivariate Adaptive Regression Splines (MARS). International Journal of Geotechnical Engineering 9:1, 79-88.
27. Liaw A., Wiener M., 2002. Classification and Regression By Random Forest, R News, 2(3).
28. Ließ, M., Glaser, B., Huwe, B., 2011. Functional Soil-Landscape Modelling To Estimate Slope Stability In A Steep Andean Mountain Forest Region. Geomorphology 132 (3–4), 287–299.
29. Magerman, D.M., 1995. Statistical Decision-Tree Models for Parsing. DecisionTree Modeling. 276–283.
30. Matteo, L.D, Bigotti, F., Ricco, R., 2009. Best-Fit Models to Estimate Modified Proctor Properties of Compacted Soil. ASCE.
31. Mehta, B., Sachan, A., 2017. Effect of Mineralogical Properties of Expansive Soil on Its Mechanical Behavior. Geotechnical Geology Engineering 35:2923–2934.
32. Omar, M., Shanabled, A., Basma, A., Barakat, S., 2003. Compaction Characteristic of Granular Soils in United Arab Emirates. Geotechnical and Geological Engineering 21: 283-295.
33. Ouedraogo, I., Defourny, P., Vanclooster, M., 2019. Application of Random Forest Regression and Comparison of Its Performance To Multiple Linear Regression In Modeling Groundwater Nitrate Concentration At The African Continent Scale. Hydrogeology Journal, 27:1081–1098. https://doi.org/10.1007/s10040-018-1900-5.
34. Özdarıcı, Ok A., Akar, Ö., Güngör, O., 2011. Rastgele Orman Sınıflandırma Yöntemi Yardımıyla Tarım Alanlarındaki Ürün Çeşitliliğinin Sınıflandırılması. TUFUAB 2011 VI. Teknik Sempozyumu, Antalya, Türkiye, ss.1-7.
35. Pal, M., 2003. Random Forest For Land Cover Classification. IEEE International Geoscience and Remote Sensing Symposium, IGARSS ‘03Proceedings, 6, 3510-3512.
36. Pal, M., 2005. Random Forest Classifier For Remote Sensing Classification. International Journal of Remote Sensing. 26(1) 217–222. https://doi.org/10.1080/01431160412331269698.
37. Pal, M., Mather, P.M., 2003. An Assessment of The Effectiveness of Decision Tree Methods For Land Cover Classification. Remote Sensing Of Environment, 86, 554-565.
38. Peng, W., Chen, J., Zhou, H., 2009. An Implementation of Decision Tree Learning Algorithm.
39. Pham, T.B., Qi, C., Ho, L.S., Thoi, T.N., Ansari, N.A., Nguyen, M.D., Nguyen, H.D., Ly, H.B., Le, H.V., Prakash, I., 2020. A Novel Hybrid Soft Computing Model Using Random Forest and Particle Swarm Optimization for Estimation of Undrained Shear Strength of Soil. Sustainability, 12, 2218; DOI:10.3390/su12062218.
40. Rastgou, M., Bayat, H., Mansoorizadeh, M., Gregory, A.S., 2020. Estimating The Soil Water Retention Curve: Comparison of Multiple Nonlinear Regression Approach and Random Forest Data Mining Technique. Computers and Electronics in Agriculture 174, 105502.
41. Sahin, K.E., 2018. Heyelan Duyarlılık Haritası İçin Adımsal Regresyona Dayalı Faktör Seçme Yönteminin Etkinliğinin Araştırılması. Harita Dergisi. Sayı 159.
42. Safavian, S.R., Landgrebe, D., 1991. A Survey of Decision Tree Classifier Methodology. IEEE Transactions On Systems Man and Cybernetics, 21, 660-674.
43. Segal, M.R., 2003. Machine Learning Benchmarks and Random Forest Regression. Center for Bioinformatics and Molecular Biostatistics, UC San Francisco. 18(3), 1-14.
44. Shukla, G., Garg, R.D., Srivastava, H.S. & Garg, K.P., 2018. An Effective Implementation and Assessment of A Random Forest Cassifier As A Soil Spatial Predictive Model. International Journal of Remote Sensing, 39:8, 2637-2669, DOI: 10.1080/01431161.2018.1430399.
45. Singh, B., Sihag, B., Singh, K., 2017. Modelling of Ompact of Water Quality On Infiltration Rate of Soil By Random Forest Regression Model. Earth Syst. Environ. 3:999–1004. DOI: 10.1007/s40808-017-0347-.
46. Singhal, S., Houston,S.L., Houston, W,N., 2015. Swell Pressure, Matric Suction, and Matric Suction Equivalent for Undisturbed Expansive Clays. Can. Geotechnical Journal 52: 356–366.
47. Sinha, S.K., Wang, M.C., 2008. Artificial Neural Network Prediction Models for Soil Compaction and Permeability. Geotechnical Geology Engineering 26:47-64.
48. Thompson, M.J., White, D.J., 2008. Estimating Compaction of Cohesive Soils from Machine Drive Power. ASCE.
49. Viji, V.K., Lissy, K.F., Sobha, C. & Benny M.A., 2013. Predictions on Compaction Characteristics of Fly Ashes Using Regression Analysis and Artificial Neural Network Analysis. International Journal of Geotechnical Engineering 7:3, 282-291.
50. Viscara Rossel, R.A., Behrens, T., 2010. Using Data Mining To Model and Interpret Soil 650 Diffuse Reflectance Spectra. Geoderma 158 (1–2), 46–54.
51. Waske B., Heinzel V., Braun M., Menz G., 2007. Random Forests for Classifying Multi-Temporal Sar Data, Proc. ‘Envisat Symposium Montreux, Switzerland, http://envisat.esa.int/envisatsymposium/proceedings/ sessions/3D3/461589wa.pdf.
52. Waske, B., Braun, M., 2009. Classifier Ensembles For Land Cover Mapping Using Multitemporal SAR Imagery. ISPRS Journal of Photogrammetry and Remote Sensing 64 (2009) 450_457.
53. Zhang, P., Yin, Z., Jin, Y., Chan, T.H.T., 2020. A Novel Hybrid Surrogate Intelligent Model For Creep Index Prediction Based On Particle Swarm Optimization and Random Forest. Engineering Geology 265, 105328.