**6. Conclusions**

The influence of the choice of the ML algorithm, sampling strategies, and data splitting for LSM is evaluated in detail using a case study from the Wayanad district in Kerala. 12 LCFs were used to develop different models using five different ML algorithms (NB, LR, KNN, RF, and SVM), two sampling strategies (point data and polygon data), and different values of k in k-fold cross validation. The results show that data splitting is least effective among the considered parameters. The performance of NB and LR are unaffected by the variation of k values, but the performance of KNN, RF, and SVM are slightly varied by k values, with the best performance at k = 8 in all cases using point data.

The performance of NB and LR did not improve with the use of a large dataset with polygon inventory. The inter dependency of parameters is a critical factor affecting the performance of these algorithms while, in the case of KNN, RF, and SVM, the performance is significantly improved with the use of polygon data. By comparing the H index values, it was observed that the landslide susceptibility maps perfectly agreed with each other in the case of 47.56% of the total area while using point data and 58.06% while using polygon data.

The results produced by KNN and RF using the polygon dataset have a very good statistical performance with very high values for accuracy and AUC. The best performing model developed using an RF algorithm and polygon dataset has an accuracy of 97.30% and an AUC of 0.99.

**Author Contributions:** Conceptualization, M.T.A., N.S. and B.P.; methodology, M.T.A. and N.S.; data curation, M.T.A. and R.L.; writing—original draft preparation, M.T.A.; writing—review and editing, B.P. and A.A.; supervision, N.S. and B.P. All authors have read and agreed to the published version of the manuscript.

**Funding:** The study is supported by the Centre for Advanced Modelling and Geospatial Information Systems (CAMGIS), University of Technology Sydney. This research was also supported by Researchers Supporting Project number RSP-2021/14, King Saud University, Riyadh, Saudi Arabia.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The publicly archived datasets used for the analysis are cited in the manuscript. The analysis has been carried out at Indian Institute of Technology Indore, and the derived data can be provided upon request to the corresponding author (Biswajeet.Pradhan@uts.edu.au).

**Acknowledgments:** The authors express their sincere gratitude to Geological Survey of India, Kerala SU, District Soil Conservation Office Wayanad and Kerala State Disaster Management Authority (KSDMA) for their support throughout the study. Authors would like to thank three anonymous reviewers for their critical reviews which helped to improve the quality of the manuscript.

**Conflicts of Interest:** Authors declare no conflict of interest.
