The performance of six neural-evolutionary classification techniques combined with multi-layer perception in two-layered cohesive slope stability analysis and failure recognition

Six population-based hybrid algorithms are applied to train the multilayer perceptron (MLP) to improve the classification accuracy, in the stability assessment. A complex problem of slope stability against failure is designed in Optum G2 software. Considering four key factors of shear strength of clayey soil, slope angle, the ratio of foundation distance from the slope to the foundation length, and the applied surcharge, the stability or failure of the proposed slope are anticipated. The provided data are used to develop the MLP combined with biogeography-based optimization (BBO), ant colony optimization (ACO), genetic algorithm (GA), evolutionary strategy (ES), particle swarm optimization (PSO) and probability-based incremental learning (PBIL). The results revealed that the BBO-MLP with the obtained area under the receiving operating characteristic curve (AUROC) of 0.995 and the classification ratio (CR) of 92.4% is the most accurate model followed by GA-MLP (AUROC = 0.960 and CR = 84.3%), PBIL-MLP (AUROC = 0.948 and CR = 79.3%), ES-MLP (AUROC = 0.879 and CR = 65.7%), PSO-MLP (AUROC = 0.878 and CR = 71.3%), and ACO-MLP (AUROC = 0.798 and CR = 60.7%).