2023 · Environmental Science
Air Pollution Abatement and Modeling
Environmental Science Regression Python
Air Pollution Abatement and Modeling
BITS Pilani | Aug 2023 – Dec 2023
Python Multivariate Regression Environmental Modeling Data Visualization
Project Overview
Developed predictive models for ambient air quality monitoring, integrating multiple data sources to forecast pollutant concentrations and inform policy recommendations.
Key Contributions
Feature Engineering: Integrated meteorological data (temperature, humidity, wind), traffic density indices, and industrial emission inventories to create comprehensive feature set
Model Development: Built multivariate regression models predicting PM2.5, NO2, and O3 concentrations achieving 85% prediction accuracy on validation data
Policy Recommendations: Proposed data-driven pollution abatement strategies based on model-identified key drivers, targeting high-impact intervention points
Technologies Used
- Languages: Python
- Methods: Multivariate Regression, Feature Selection
- Pollutants Modeled: PM2.5, NO2, O3
- Data Sources: CPCB, Weather APIs
Key Results
| Metric | Value |
|---|---|
| Prediction Accuracy | 85% |
| Pollutants Modeled | 3 (PM2.5, NO2, O3) |
| Feature Categories | Meteorological, Traffic, Industrial |