**3. Methodology**

This study aims at evaluating the uncertainties in LSM using ML by adopting different ML algorithms, sampling strategies, and train to test ratios. The first step was the preparation of the dataset, starting from the landslide inventory. The data has to be preprocessed before using it for training and testing. Five different ML approaches were used in this study for comparison.

#### *3.1. Machine Learning Algorithms*

Data-based methods are often used to solve real-world problems when the knowledge of the theoretical part is limited and the data is of a large size [19]. Being a non-linear problem, ML models are highly suitable for LSM. The algorithms can learn the association between the occurrence or non-occurrence of landslides and the LCFs using the landslide and no landslide points used for training. Five different ML algorithms are considered in this study, which are explained as follows:
