Landslide susceptibility assessment (LSA) is fundamental for managing landslide geological disasters. This study presents a deep learning approach (DNN-MSFM) designed to enhance LSA modeling, particularly addressing limitations caused by the unbalanced distribution of data samples in applied datasets. DNN-MSFM approach combines a deep neural network (DNN) and a mean squared false misclassification loss function (MSFM) to handle unbalanced samples from the algorithmic perspective. The model's performance was evaluated on an unbalanced dataset containing mapping units' records of 293 landslide samples and 653 non-landslide samples from the Baota District, China. Its effectiveness was assessed through statistical metrics and compared against DNN and Support Vector Machine (SVM) basic models. The results demonstrated a significant performance enhancement from the DNN-MSFM (OverallAccuracy = 0.889 and area under the receiver operating characteristic curve (AUC) = 0.84), indicating its effectiveness in learning the underlying landslide susceptibility features and demonstrating its ability to provide improved predictions even in areas with unbalanced landslide samples. Moreover, the study emphasizes the importance of considering balanced loss functions in training DNN under various imbalance degrees and contributes to expanding the applicability of DNN in LSA modeling. Also, this study builds a foundation for further enhancements of deep learning methods for geological disaster assessments.

Mountainous open-pit mines are highly susceptible to landslides, yet quantitative risk assessment remains a challenge. This study aims to develop and validate a quantitative landslide risk assessment model by integrating multi-source data to enhance hazard identification in these complex environments. Taking the Dexing Copper Mine as a case study, we used Small Baseline Subset InSAR (SBAS-InSAR) to derive surface deformation rates. This deformation data was integrated with geological and topographical factors within a Geographic Information System (GIS), using an information value model combined with weighting from the Analytic Hierarchy Process (AHP) and Entropy Weight Method (EWM) to generate a comprehensive landslide risk map. The results show that 3860 potential landslide points were identified, with deformation rates ranging from −338.74 to 80.61 mm/a. High and very high-risk zones were primarily concentrated around the Fujiawu and Zhujiawu dump sites, and the model’s performance was validated with a high degree of accuracy, achieving an Area Under the Curve (AUC) value of 0.871. This study demonstrates that the integration of multi-source data provides a robust and effective approach for quantitative landslide risk assessment in mountainous mining areas. The proposed framework can serve as a valuable tool for targeted disaster prevention and management.