Introduction:
Air pollution poses a significant threat to public health globally. This project tackles the air pollution challenge in a heavily polluted city in Italy by focusing on two key tasks:
1. Predicting Carbon Monoxide (CO) Concentration:
- We develop a robust model to predict the hourly average CO concentration using machine learning regression techniques based on environmental data and meteorological factors.
- This prediction allows for informed decision-making related to public health and air quality management.
2. Defining an Air Quality Index (AQI):
- We create a user-friendly AQI specifically tailored for the city, providing a clear and concise assessment of overall air quality based on multiple pollutants.
- This AQI empowers citizens to make informed decisions about their health and activities based on real-time air quality information.
Data and Methodology:
- We leverage a dataset containing hourly average concentrations of various air pollutants and meteorological factors within the city.
- For CO prediction, we employ feature engineering and advanced regression algorithms like XGBoost to achieve high accuracy.
- For AQI development, we analyze the relationships between individual pollutants and their combined impact on health, establishing a scoring system with clear air quality categories.
Results and Impact:
- The CO prediction model achieves high accuracy and generalizability, enabling reliable forecasting of air quality.
- The user-friendly AQI effectively communicates air quality to the public, promoting informed decision-making and potentially improving health outcomes.
- Our work contributes to improved air quality management strategies and public awareness campaigns, ultimately leading to a healthier living environment for the city's residents.
Future Work:
- We plan to integrate real-time data into the CO prediction model for continuous updates and improved air quality monitoring.
- We aim to refine and validate the AQI based on user feedback and expert evaluation, ensuring its effectiveness and ease of understanding.
- We envision developing user-friendly interfaces for visualizing CO predictions and the AQI, further empowering citizens to take action for cleaner air.
Collaboration:
We welcome collaboration with stakeholders like city authorities and environmental organizations to implement the CO prediction model and AQI for air quality management and public awareness campaigns.
By contributing to cleaner air and improved public health, this project aims to leave a positive impact on the polluted Italian city and its residents.
This README file provides a concise overview of the project. For a more detailed description, please refer to the individual notebooks and reports within the repository.