Landslide Susceptibility Prediction in Attappadi, Kerala Using Machine Learning and GIS: A Research Paper / Paper Presentation

Sreekanth V

Abstract: Landslides are major natural hazards in mountainous regions and are frequently associated with intense rainfall, steep slopes, weathered materials, drainage conditions, land-use changes and human activities. Attappadi in Palakkad district, Kerala, lies in the Western Ghats and contains rugged terrain where rainfall-induced slope instability can become a serious hazard. This paper proposes a machine-learning and Geographic Information System (GIS) based framework for landslide susceptibility prediction in Attappadi. The proposed framework integrates terrain, hydrological, environmental and anthropogenic conditioning factors such as elevation, slope, aspect, rainfall, soil, land-use/land-cover, vegetation, drainage, distance from roads and distance from streams. Historical landslide locations are used to construct a landslide inventory and create labelled training and testing datasets. Machine-learning algorithms such as Random Forest, Support Vector Machine and XGBoost can be trained and compared using accuracy, precision, recall, F1-score and ROC-AUC. The trained model can generate a spatial susceptibility index and classify the Attappadi landscape into Very Low, Low, Moderate, High and Very High susceptibility classes. The proposed system is intended as a decision-support tool for hazard assessment, land-use planning and disaster preparedness. It does not replace official warnings or site-specific geotechnical investigations.

Keywords: Landslide susceptibility, Attappadi, Kerala, Machine Learning, Random Forest, GIS, Remote Sensing, Western Ghats

How to Cite?: Sreekanth V, "Landslide Susceptibility Prediction in Attappadi, Kerala Using Machine Learning and GIS: A Research Paper / Paper Presentation", Volume 15 Issue 9, September 2026, International Journal of Science and Research (IJSR), Pages: 224-226, https://www.ijsr.net/getabstract.php?paperid=SR26901211536, DOI: https://dx.doi.org/10.21275/SR26901211536

Download Citation: APA | MLA | BibTeX | EndNote | RefMan