Open AccessA multiple model machine learning approach for soil classification from cone penetration test data
Author(s) -
Lucas Orbolato Carvalho,
Dimas Betioli Ribeiro
Publication year - 2021
Publication title -
soils and rocks/soils and rocks
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.161
H-Index - 10
eISSN - 2675-5475
pISSN - 1980-9743
DOI - 10.28927/sr.2021.072121
Subject(s) - cone penetration test , ensemble learning , computer science , machine learning , artificial intelligence , penetration test , test data , data mining , geotechnical engineering , geology , subgrade , programming language
The most popular methods for soil classification from cone penetration test (CPT) data are based on examining two-dimensional charts. In the last years, several authors have dedicated efforts on replicating and discussing these methods using machine learning techniques. Nonetheless, most of them apply few techniques, include only one dataset and do not explore more than three input features. This work circumvents these issues by: (i) comparing five different machine learning techniques, which are also combined in an ensemble; (ii) using three distinct CPT datasets, one composed of 111 soundings from different countries, one composed of 38 soundings with information of soil age and the third composed of 64 soundings taken from the city of São Paulo, Brazil; and (iii) testing combinations of five input features. Results show that, in most cases, the ensemble of multiple models achieves better predictive performance than any technique isolated. Accuracies close to the maximum were obtained in some cases without the need of pore pressure information, which is costly to measure in geotechnical practice.