This study develops a landslide susceptibility map using a data-driven approach and taking the district Muzaffarabad, northwest Himalayas, Pakistan, as a case study. The Muzaffarabad district was severely affected by the 2005 Kashmir earthquake triggered landslides. To manage these landslides, it is necessary to collect data and information about landslides of the Muzaffarabad district. Therefore, a landslide inventory was prepared by using remote sensing imageries, and 459 landslides were identified. These landslides were divided into training and validation samples ($70\text{:}30\%$ ). About 321 known landslide points were used as training sites in combination of nine causative factor rasters to produce a landslide susceptibility map of the area. About 20% of all training sites were verified through field mapping, and an accuracy of 77% was achieved. The selected causative layers include landcover, lithology, slope angle, elevation, distance to drainage, distance to faults, distance to roads, aspect, and slope curvature. The landslide susceptibility map was classified into low, moderate, high, and very high susceptible zones. The validity of the model was assessed with a success rate curve (SRC), whereas prediction efficiency was analyzed through a prediction rate curve (PRC). The validation results reveal fine compatibility between landslides and the obtained posterior probability model of the area. The efficiency of classification is 89%, and efficiency of prediction is 86.2%. To generate posterior probability, the cumulative area posterior probability (CAPP) curve was used to reclassify the continuous posterior probability scale map into four classes. According to the results of weights calculated through Arc-SDM, landcover, lithology, and slope angle are the three key parameters to determine the occurrences of landslides in the study area. To implement the landslide management strategies, the developed landslide susceptibility map can be used by the concerned agencies.

Several devastating landslides have occurred in the NW Himalayas, which has prompted several researchers to strive for improvement in landslide susceptibility modelling (LSM) methodologies. This research analyzes the effectiveness of alternative landslide partitioning techniques on LSM in the landslide-prone district, Muzaffarabad, Pakistan. We developed a landslide inventory of 961 landslides and then traditionally divided it into training (672; 70%) and testing (289; 30%) samples. These training samples (672) are processed by the Average Nearest Neighbour Index (ANNI) method to estimate the spatial pattern of landslides in nature. The results provide an ANNI ratio of 0.672 confirming that the landslides distribution pattern is cluster in the complex Himalayan terrain of Muzaffarabad. Among 672, the majority of landslides (529; 79%) depict cluster behaviour, while 189 landslides (21%) depict random behaviour. To evaluate the effectiveness of landslide cluster samples in prediction, five machine learning algorithms (MLAs), that is, K-Nearest Neighbour (KNN), Naïve Bayes (NB), Random Forest (RF), Extreme Gradient Boosting (XGBoost) and Logistic Regression (LR) using proposed cluster (529) and traditional (672) random training samples along with 17 geo-environmental factors are executed. However, testing samples (289; 30%) separated at the initial stage remained the same to check the model’s effectiveness. The areas under the curve (AUC-ROC), sensitivity, specificity, Kappa index and accuracy (ACC) have been used to evaluate the MLA’s performances. An alternative partitioning technique (cluster) shows the highest predictive power with AUC-ROC values ranging from 0.96 to 0.86, Kappa index ranges from 0.76 to 0.60 and ACC ranges from 0.90 to 0.83. Conversely, the random partitioning approach performs less well with AUC-ROC values ranging from 0.95 to 0.83, Kappa index ranges from 0.70 to 0.49 and ACC ranges from 0.87 to 0.80. In comparison, the RF cluster sampling-based model outperforms the other models and their counterparts. The RF model achieved the highest accuracy (0.902), highest AUC (0.962) and highest Kappa index (0.755) followed by XGboost having ACC (0.885), AUC (0.95) and Kappa index (0.724) employing proposed cluster training samples. However, traditional random training samples yield comparatively low ACC of RF (0.868) and XGboost (0.862). These results confirm that cluster training sampling performs well in obtaining reliable and precise LSMs for complex Himalayan terrain. Although random landslide partitioning for training datasets is seldom utilized in LSM, this study highlights that cluster partitioning for landslide training datasets might be a realistic and reliable approach.