Coastguard Ai: A Context-Aware Chatbot for Disaster Preparedness and Response in Philippines Coastal Communities
Year: 2026
Type: conference-paper
Abstract: The Philippines is very vulnerable to natural disasters, as it is located within the Pacific Ring of Fire and the Western Pacific typhoon region, with coastal populations being particularly affected. In this work, we propose the design and development of a context-aware disaster response chatbot empowered by a pre-trained Large Language Model (LLM) with Retrieval-Augmented Generation (RAG). The technology aims to offer precise, localised and immediately available disaster-related information using the cloud-based online platform. The chatbot enables multilingual conversation in 23 languages utilised by local and international users in the Philippines to enhance accessibility and diversity. The chatbot provides important disaster-related information including evacuation orders, weather alerts, safety precautions, emergency contacts, and resource locations using an interactive conversational interface. The system employs validated disaster datasets in the RAG framework to improve the accuracy and contextual appropriateness of its replies. The study shows the potential of AI-enabled communication platforms to enhance catastrophe preparedness, improve risk communication, and build community resilience. The proposed system supports the development of scalable and inclusive catastrophe management systems for climate adaption, public safety and sustainable development efforts.
Institution: Adamson University
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: gold
OpenAlex ID: W7202118156