Earthquake risk assessment in NE India using deep learning and geospatial analysis

Abstract Earthquake prediction is currently the most crucial task required for probability, hazard, risk mapping, and mitigation purposes. attracts researchers' attention from both academia industries. Traditionally, assessment approaches have used various traditional machine learning models. However, deep techniques been rarely tested earthquake probability mapping. Therefore, this study develops a convolutional neural network (CNN) model in NE India. Then conducts vulnerability using analytical hierarchy process (AHP), Venn's intersection theory integrated A of classification was performed which predicts magnitudes more than 4 Mw that considers nine indicators. Prediction results intensity variation were then hazard respectively. Finally, map produced by multiplying vulnerability, coping capacity. The prepared six vulnerable factors, capacity estimated number hospitals associated variables, including budget available disaster management. CNN distribution robust technique provides good accuracy. Results show superior to other algorithms, completed with an accuracy 0.94, precision 0.98, recall 0.85, F1 score 0.91. These indicators total area (21,412.94 km2), (480.98 (34,586.10 km2) estimated.