Advisers

Dr. Remedios P. Magnaye

Discipline

Engineering, Information, And Communication Technology

Abstract

This study investigates the utilization and comparative performance of artificial intelligence (AI) algorithms in disaster response, specifically in addressing typhoons, soil erosion, floods, and hurricanes. By reviewing published academic literature and datasets, this study identifies commonly used AI models, such as Random Forest (RF), Artificial Neural Networks (ANNs), Support Vector Machines (SVMs), Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM) models, and Convolutional Neural Networks (CNNs). These algorithms facilitated early warning, risk evaluation, disaster response, and structural assessment by analyzing historical and real-time data. Deep learning models, including CNNs, RNNs, and LSTM models, excelled in meteorological prediction for typhoons and hurricanes due to their high accuracy, but they required extensive computational resources and large datasets. In contrast, machine learning models, such as RF and K-Nearest Neighbors (KNN), were cost-effective and efficient in managing structured data, making them suitable for flood and soil erosion prediction. The study highlighted that RF and KNN offered the lowest computational costs, while deep learning models required higher investment. An action plan was proposed, focusing on six key areas: early warning, risk assessment, response, education, structural monitoring, and recovery. This plan incorporated predictive modeling, resource management, and community awareness to strengthen disaster management frameworks. The study concluded that integrating AI-driven decision support with government policy and public education could enhance disaster preparedness and mitigation. Future research should explore hybrid AI models and the integration of AI with IoT and remote sensing to improve real-time disaster monitoring and response effectiveness.

Keywords

artificial intelligence, ai algorithms, disaster response, early warning, risk assessment, disaster preparedness