AI-Powered Forecasting: Revolutionizing Natural Disaster Prediction and Response Optimization

Gankidi, Sai Charan Reddy (2024) AI-Powered Forecasting: Revolutionizing Natural Disaster Prediction and Response Optimization. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (512kB) | Preview
Preview	PDF (Configuration Manual) Download (515kB) | Preview

Abstract

The current frequency and intensification of natural disasters, influenced by climate change, remains a significant threat to global societies; human and material losses. In view of the generally recognized need for upgraded disaster prediction and improved emergency procedures, this research aims, primarily, at application of the crucial artificial intelligence (AI) methods in the context of natural disasters prediction. The research uses ML models for numerical data and DL models, such as CNN, VGG16, and ResNet50, for image-based disaster categorization. To aid practical deployment of disaster prediction and make it real-time, an interface using Gradio is designed.

The methodological approach starts with data pre-processing where data is cleaned from anomalies, outliers and some features are engineered. During exploratory data analysis, important characteristics, including cyclical patterns and differences in the incidence of disasters by region, are identified. The analyses of the chosen model demonstrate the increases in accuracy and, therefore, the enhanced results in the classification of the disaster types and the prediction of the program declarations. The interface developed with the help of Gradio improves the available adjustments and ensures real-time operability.

The implications of this research are to show the utility of AI in disaster management if the objective is a reduction in losses. The findings lay the path for future development application in real-time integration of IoT systems and other larger data sets, thereby repositioning AI as the central tool in handling global disasters.

Item Type:	Thesis (Masters)
Supervisors:	Name Email -, - UNSPECIFIED
Subjects:	G Geography. Anthropology. Recreation > GE Environmental Sciences Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Artificial Intelligence
Depositing User:	Ciara O'Brien
Date Deposited:	19 Jun 2025 15:34
Last Modified:	19 Jun 2025 15:34
URI:	https://norma.ncirl.ie/id/eprint/7945

Actions (login required)

View Item