This journal article provides a susceptibility prediction of Landslides. A semi-supervised model, SS-PSO-ELM, was proposed for landslide susceptibility prediction, addressing challenges in expanding landslide samples and increasing accuracy. It combines Density Peak Clustering, Frequency Ratio, and Random Forest models to expand and categorize landslide data. The SS-PSO-ELM model outperforms other models with an AUC of 0.893 and RMSE of 0.370, indicating its effectiveness in predicting landslide susceptibility in Fu’an City, Fujian Province. This approach offers a new research direction for landslide prediction.
International Journal of Geo-Information