Machine Learning: Alternative or Supplementary Method for​ Short-Term Weather Forecasting​

Machine Learning: Alternative or Supplementary Method for​ Short-Term Weather Forecasting​ 

26 November 2021 | 1:00 – 2:00 PM

via Zoom​ 

[ACCESS PRESS KIT HERE]


Overview

In line with this year’s National Science and Technology Celebration, the DOST-ASTI will lead discussions on the applicability of machine learning in weather forecasting as a supplement or an alternative to the traditional methods. This event will introduce the field of artificial intelligence (AI) and its pervasiveness in a practical and accessible manner, that assumes no prior knowledge of the subject. The speaker will also elaborate on different ideas and methodologies in machine learning as well as a few concepts in weather forecasting. Recent research studies that validate the accuracy and reliability of deep learning methods for short-term forecasting will also be demonstrated. The last part of the talk will zero in on the real-world application of the theoretical underpinnings through a discussion about the ULAT project.

Schedule of Activities

1:00 – 1:05 PM  Welcome Message 
1:05 – 1:40 PM  Talk
  • Weather Forecasting Technologies​  
  • What is Machine Learning?​  
  • Machine Learning applied to weather forecasting​  
  • The ULAT project: Understanding Lightning and Thunderstorms
1:40 – 2:00 PM  Q&A Session​ 

Speakers

Elmer C. Peramo​  

Senior Science Research Specialist 

Elmer C. Peramo holds a Bachelor of Science in computer science degree in Mondriaan Aura College and a master’s in Electrical Engineering at the University of the Philippines–Diliman. He’s currently a Senior Science Research Specialist at the Advanced Science and Technology Institute (DOST-ASTI) and a PhD student in Computer Science at the De La Salle University. His research focus is on artificial intelligence and its applications in agriculture, business, education, health, disaster risk reduction, and urban computing. He’s a technical researcher of the ULAT Project—a successful collaboration among PAGASA, UP-IESM, Hokkaido University, and JICA under the initiative of Japan’s Science and Technology Research Partnership for Sustainable Development (SATREPS) Program that aims to observe the country’s weather behavior and study lightning, torrential rainfall, and thunderstorm occurrences to eventually enable short-term forecasting.