<![CDATA[The primary objective of this research is to examine the viability of simplified regularized regression models in predicting the slope safety factor of road embankments. The methodology involves developing and comparing several regularized linear regressions against conventional methods. A total of 276 data points are collected from the literature, and 70% of these are utilized for model training, while 30% are employed for testing. The findings indicate that these models yield results better than established approaches, with Stochastic Gradient Descent and Bayesian Ridge achieving strong performances. This study provides an alternative technique that offers rapid and manually solvable equations, thus enhancing practical adaptability for routine professional tasks. The novelty lies in bridging the gap between traditional finite element-based investigations and emerging data-driven methods, demonstrating that regularized regression can be both simple and sufficiently accurate. Overall, the study outcomes emphasize the significance of these advanced yet computationally light models for road embankment stability assessments, presenting a valuable and time-efficient tool for practitioners.]]>
Copyrights © 2025
