Related Experiment Video
Updated: Oct 15, 2025
Asthma Detection Research Based on Voice Signal Processing and Machine Learning
Published on: July 22, 2025
Development and application of an automated air quality forecasting system based on machine learning
Huabing Ke1, Sunling Gong2, Jianjun He2
1Climate and Weather Disasters Collaborative Innovation Center, Nanjing University of Information Science & Technology, Nanjing 210044, China; State Key Laboratory of Severe Weather & Key Laboratory of Atmospheric Chemistry of CMA, Chinese Academy of Meteorological Sciences, Beijing 100081, China.
An automated machine learning system provides accurate daily air quality forecasts for key pollutants like PM2.5 and ozone. This advanced system outperforms traditional numerical models, offering a promising tool for environmental meteorology.
Area of Science:
- Environmental Science
- Atmospheric Science
- Computer Science
Background:
- Air quality is a major societal concern, driving advancements in forecasting technology.
- Accurate air quality prediction is crucial for public health and environmental management.
Purpose of the Study:
- To develop and evaluate an automated machine learning system for daily air quality forecasting.
- To forecast concentrations of six common pollutants (PM2.5, PM10, SO2, NO2, O3, CO) and pollution levels.
- To automatically optimize model selection and hyperparameters without manual intervention.
Main Methods:
- Developed an automated system integrating five machine learning models and a stacked generalization ensemble model.
- Utilized a knowledge base with meteorological, pollutant concentration, emission, and reanalysis data.
- Applied the system to five years of data (2015-2019) from seven major Chinese cities.
Main Results:
- The automated system achieved satisfactory forecasting performance across seven evaluation criteria.
- Forecasts for the next three days demonstrated the system's effectiveness.
- The system's performance surpassed that of most numerical models.
Conclusions:
- The developed automated air quality forecasting system shows significant potential for practical application.
- This machine learning approach offers a robust alternative to traditional forecasting methods.
- The system contributes to advancing environmental meteorology and air quality management.
