Investigating the Soil Unconfined Compressive Strength Based on Laser-Induced Breakdown Spectroscopy Emission Intensities and Machine Learning Techniques

Laser-induced breakdown spectroscopy (LIBS) is a remarkable elemental identification and quantification technique used in multiple sectors, including science, engineering, medicine. Machine learning techniques have recently sparked widespread interest the development of calibration-free LIBS due to their ability generate defined pattern for complex systems. In geotechnical understanding soil mechanics relation applications paramount importance. The knowledge unconfined compressive strength (UCS) enables engineers identify behaviors particular propose effective solutions given problems. However, experimental involved measurements UCS are incredibly expensive time-consuming. this work, we develop pioneering estimate using artificial intelligent methods based on spectra obtained from system. Decision tree regression (DTR) support vector learners were initially employed, consequently, adaptive boosting method was applied improve performance two single learners. prediction power established models determined standard evaluation metrics such as root-mean-square error, CC between predicted actual values, mean absolute R2 score. Our results revealed that boosted DTR exhibited highest coefficient correlation 99.52% an value 99.03% during testing phase. To validate models, values soils stabilized with lime cement optimum degree accuracy, confirming models’ suitability generalization investigations.