August 4, 2025

Feature Attribution-Driven Flood Susceptibility Assessment: an Integrated Gradients Approach in Seti Gandaki River Basin

View Full Paper

Ask AI

Bookmark

Authors

BKBiswash KaphleTribhuvan University AAAayush AdhikariTribhuvan University Teaching Hospital AKAkinchan KafleUniversity of Alberta

Discussion

Member takes

Implication

Neural network analyzes flood susceptibility in Seti Gandaki watershed, suggesting enhanced predictive capabilities through innovative geospatial techniques.

Key Points

The framework assesses flood susceptibility using deep learning, achieving over 98% accuracy and significant precision in predictions.
Feature attribution reveals Topographic Wetness Index, distance to rivers, and rainfall as key factors in flood risk.
A comprehensive geospatial approach integrates data preprocessing and model training for reliable outcomes in flood prediction.
Comparative results show only 8.52% agreement with traditional AHP methods, highlighting the advancements of modern techniques.

Cite This Study

Kaphle et al. (2025) studied this question.

synapsesocial.com/papers/689a0f93e6551bb0af8d0fbc https://doi.org/10.21203/rs.3.rs-7052855/v1

View Full Paper

Ask AI

Bookmark

Implication

Neural network analyzes flood susceptibility in Seti Gandaki watershed, suggesting enhanced predictive capabilities through innovative geospatial techniques.

Feature Attribution-Driven Flood Susceptibility Assessment: an Integrated Gradients Approach in Seti Gandaki River Basin

Discussion

Member takes

Implication

Key Points

Cite This Study

Feature Attribution-Driven Flood Susceptibility Assessment: an Integrated Gradients Approach in Seti Gandaki River Basin

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider