A machine learning-based model to estimate PM2.5 concentration levels in Delhi's atmosphere
Saurabh Kumar1, Shweta Mishra1, Sunil Kumar Singh1
1Department of Computer Science & Information Technology, Mahatma Gandhi Central University, Bihar, India.
Heliyon
|December 11, 2020
Summary
Delhi
Area of Science:
- Environmental Science
- Atmospheric Science
- Public Health
Background:
- Delhi's air quality has been hazardous for years, leading to respiratory illnesses.
- High concentrations of fine particulate matter (PM2.5) are a primary cause.
- Accurate PM2.5 forecasting is crucial for public health interventions.
Purpose of the Study:
- To forecast hourly PM2.5 concentration levels across Delhi.
- To develop a predictive model using time series analysis and regression.
- To inform public health strategies and safety measures.
Main Methods:
- Utilized time series analysis and regression techniques.
- Incorporated atmospheric factors like wind speed, temperature, and pressure.
- Developed a regression model using Extra-Trees and AdaBoost for enhanced prediction.
Main Results:
- The proposed model accurately forecasts hourly PM2.5 levels.
- Experimental results demonstrate the model's efficacy compared to existing methods.
- The model provides valuable data for air quality management.
Conclusions:
- The developed model offers a reliable method for PM2.5 forecasting in Delhi.
- Accurate forecasting can aid in mitigating health risks associated with air pollution.
- This research contributes to improving public health outcomes in polluted urban environments.
