Proceedings of the
8th International Symposium on Geotechnical Safety and Risk (ISGSR)
14 – 16 December 2022, Newcastle, Australia
Editors: Jinsong Huang, D.V. Griffiths, Shui-Hua Jiang, Anna Giacomini, Richard Kelly
doi:10.3850/978-981-18-5182-7_10-005-cd
3D Underground Stratification Using GLASSO
Graduate School of Environmental and Life Science, Okayama University, 3-1-1 Tsushima naka, Kita-ku, Okayama 700-8530, Japan.
ABSTRACT
This paper presents application examples of Geotechnical lasso (Glasso) in three-dimensional underground stratification of actual sites, Hollywood (South Carolina, USA) (Stuedlein et al. 2016) and Baytown (Texas, USA). Glasso is a data-driven site characterization method for estimating trends and detecting layer boundaries consistently and is based on a sparse machine learning method called least absolute shrinkage and selection operator (lasso). Cone penetration test (CPT) data are available in both sites, and they were converted to the soil behavior type (SBT) index for the underground stratification. The performance of Glasso in underground stratification was evaluated with validation set based on two performance metrics, root-mean-square error (RMSE) of Ic and identification ratio (IR) of SBT. Glasso is capable of detecting layer boundaries without the need to choose basis functions, and this is a notable advantage of Glasso in underground stratification.
Keywords: Soil stratification, Geotechnical lasso, Cone penetration test
