Answers are grounded in this paper and cite their sources.
Accurate slope stability classification is essential for preliminary risk assessment, while Random Forest (RF) performance depends strongly on appropriate hyperparameter selection. This study develops an Improved Pelican Optimization Algorithm–Random Forest (IPOA-RF) model using 898 slope cases and six physically interpretable variables, including unit weight (γ), cohesion (c), internal friction angle (φ), slope angle (β), slope height (H), and pore water ratio (r u ). The proposed IPOA integrates sine-chaotic initialization, an improved sine–cosine search strategy, and Levy flight mechanisms to optimize three RF hyperparameters, including the number of trees, the number of predictors at each split, and the minimum leaf-node size. Benchmark-function experiments showed that IPOA achieved competitive optimization accuracy, convergence performance, and stability compared with six representative swarm intelligence algorithms. For slope stability prediction, IPOA-RF was compared with six metaheuristic optimization-based RF models and two non-metaheuristic RF optimization baselines using a stratified nested ten-fold cross-validation framework. IPOA-RF achieved an outer-test accuracy of 89.65% ± 4.49%, an F1-score of 0.9056 ± 0.0415, and an AUC of 0.9609 ± 0.0212. Statistical analyses indicated that IPOA-RF exhibited competitive predictive performance under the current experimental conditions. SHAP analysis identified unit weight (γ) and cohesion (c) as influential variables and revealed nonlinear relationships among input parameters. These results suggest that IPOA provides an effective strategy for RF hyperparameter optimization and supports data-driven preliminary slope stability classification. Further validation using independent slope cases under diverse geological and hydrological conditions is required.
No takes yet. Share an insight, caveat, or question.
Junfeng et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: