AIRCoDe Publishes Deep Learning-Based River Flood Forecasting Research in IEEE Xplore
The AI Research Center for Community Development (AIRCoDe) of Camarines Sur Polytechnic Colleges (CSPC) proudly announces the publication of its research paper titled “Project Apaw: Spatiotemporal Forecasting of River Flood Using Deep Learning” in the IEEE Xplore Digital Library as part of the proceedings of the 2023 IEEE 15th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment, and Management (HNICEM).
The AI Research Center for Community Development (AIRCoDe) of Camarines Sur Polytechnic Colleges (CSPC) proudly announces the publication of its research paper titled “Project Apaw: Spatiotemporal Forecasting of River Flood Using Deep Learning” in the IEEE Xplore Digital Library as part of the proceedings of the 2023 IEEE 15th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment, and Management (HNICEM).
Authored by Joseph Jessie S. Oñate, Challiz D. Omorog, Rosel O. Onesa, Kaela Marie N. Fortuno, Brenda D. Benosa, Keith Marlon R. Tabal, Ian P. Benitez, Philip Alger M. Serrano, Freddie B. Prianes, and Ichelle F. Baluis, the study explores the application of deep learning techniques in forecasting river flood risks, particularly in vulnerable and low-lying communities such as the Municipality of Nabua, Philippines.
The research involved the development and deployment of a localized weather monitoring system composed of an Indoor Control Unit (ICU) and an Outdoor Control Unit (OCU). These units collect essential environmental data, including temperature, humidity, rainfall, water level, and wind speed, and transmit the gathered information to a web server through RS485 Modbus communication.
Using the collected sensor data, the researchers developed a Long Short-Term Memory (LSTM) model to identify patterns and relationships in environmental conditions and predict potential river flooding. The resulting spatiotemporal forecasting model demonstrated promising performance, achieving a Root Mean Square Error (RMSE) of 0.0101 and a Mean Absolute Error (MAE) of 0.0097.
The findings highlight the potential of deep learning in strengthening localized flood forecasting and supporting proactive disaster risk reduction and management. The study also emphasizes the importance of improving data availability and quality, incorporating spatial and climate-related variables, and expanding long-term data collection to enhance the reliability and applicability of AI-driven flood prediction systems.
The publication marks another milestone in AIRCoDe's commitment to advancing artificial intelligence research and developing innovative, data-driven solutions to address community challenges, particularly in disaster preparedness and environmental management.
This research was made possible through the generous support of the Department of Science and Technology – Philippine Council for Industry, Energy and Emerging Technology Research and Development (DOST-PCIEERD) under its Institutional Development Program (IDP), as part of the project titled "Establishment of the AI Research Center for Community Development."
Publication Details:
Publication: 2023 IEEE 15th International Conference on HNICEM
Repository: IEEE Xplore Digital Library
DOI: https://doi.org/10.1109/HNICEM60674.2023.10589082
Funding Agency: DOST-PCIEERD
Congratulations to the entire research team for this significant academic contribution!