The aim of this study is to assess and compare two clustering algorithms, namely, the Wave-cluster and Density Based Spatial Clustering of Application with Noise (DBSCAN) algorithms, for landslide susceptibility assessment in Baota District, China. Based on historical reports, interpretation of aerial photographs and field survey reports 293 were identified in the study area. Seven factors associated with landslide occurrence were prepared and used as inputs into the models for landslide susceptibility modelling. The models divided the study area grids into 478 and 465 subclasses respectively, which were further grouped into five susceptibility levels using K-means algorithm, based on the landslide density values of the obtained subclasses. The resulting two models were validated and compared using area under the receiver operating characteristics (AUC-ROC) curve and statistical measures including sensitivity, specificity and accuracy. The analysis demonstrated that the Wave-cluster model outperformed the DBSCAN model with sensitivity, specificity and accuracy values of 88.05%, 89.2% and 88.54% respectively, while DBSCAN had sensitivity, specificity and accuracy values of 86.69%, 86.38% and 86.56% respectively. In addition, the Wave-cluster model had higher value AUC of 0.823 than the DBSCAN model (0.808). This indicates that, the landslide susceptibility maps (LSMs) generated in this study had good performance for assessing landslide susceptibility in the study area and can be useful tools in assessing landslides, as the foundation for preventing and reducing landslide consequences for suitable social and economic development. Moreover, this is the first time these models are assessed and compared, especially in landslide susceptibility assessment.