What is it about?
his paper presents a new method for predicting ozone levels in Marrakech, Morocco, using an advanced deep-learning model called a Transformer. Ozone, a harmful pollutant, can cause respiratory and environmental damage. The study uses historical ozone and weather data to train the model, predicting ozone concentrations over multiple time horizons—from one hour up to 24 hours ahead. The goal is to improve air quality management in Morocco by accurately forecasting ozone levels and helping to prevent its harmful impacts on public health.
Featured Image
Photo by Max Brown on Unsplash
Why is it important?
This work is important because it addresses the growing issue of air pollution in rapidly urbanizing cities like Marrakech. Ozone, a major pollutant, poses severe health risks. By introducing Transformer models—typically used in language processing—into air quality forecasting, this research improves the accuracy of predictions compared to traditional models. The ability to predict ozone concentrations up to 24 hours in advance can empower authorities to implement timely interventions, thereby safeguarding public health and the environment.
Perspectives
Writing this paper was a rewarding experience, as it allowed us to explore cutting-edge technology, the Transformer model, and apply it to real-world environmental challenges. Working on improving air quality predictions in Marrakech, a city with increasing pollution, felt impactful. We hope that this research will inspire further applications of advanced AI techniques in environmental monitoring, not just in Morocco but around the world.
BEKKAR Abdellatif
Faculty of Sciences and Technologies, Mohammedia (FSTM)
Read the Original
This page is a summary of: Transformer-Based Model for Multi-Horizon Forecasting Ozone in Marrakech city, Morocco, November 2023, Institute of Electrical & Electronics Engineers (IEEE),
DOI: 10.1109/sita60746.2023.10373756.
You can read the full text:
Contributors
The following have contributed to this page
