Air Quality Analysis and Prediction Model

COVID-19's Impact on PM2.5 Concentrations and Climate Change: A Machine Learning Approach

Overview

This repository contains a machine learning model developed using Python to analyze the impact of COVID-19 on PM2.5 concentrations and climate change. The model utilizes time series analysis to predict PM2.5 concentrations and explores the potential of machine learning in addressing climate change.

Background

The COVID-19 pandemic has inadvertently provided a unique opportunity to reassess our priorities and strive for a more environmentally conscious future. This project aims to investigate the impact of COVID-19 on PM2.5 concentrations in India and explore the role of machine learning in climate change mitigation.

Methodology

• Time series analysis was used to analyze the PM2.5 concentration data in India during the COVID-19 pandemic.
• SARIMA and Holt-Winters machine learning models were developed using Python to predict PM2.5 concentrations based on historical data.
• The Holt-Winters Model was concluded to be the better predictor as the SARIMA model failed to take into consideration the unexpected increase in PM2.5 concentrations.

Key Findings

• The analysis revealed a significant decrease in PM2.5 concentrations in India during the COVID-19 pandemic.
• The machine learning model demonstrated the potential to predict PM2.5 concentrations with high accuracy.
• The project highlights the importance of sustained efforts to reduce emissions and transition towards a more sustainable future.
• The COVID-19 Lockdown impact on climate change and air quality can be expressed clearly based on the analysis in the project.
• The upliftment of lockdown in 2022 can also be seen to have had an exponential impact on the PM2.5 concentration, indicating poor air quality and higher climate change.

Machine Learning Model

• The machine learning model was developed using Python and utilizes time series analysis to predict PM2.5 concentrations.
• The model can be used to forecast PM2.5 concentrations and support data-driven decision making for climate change mitigation.

Future Directions

• The integration of machine learning into climate change research and policy can enhance our ability to adapt to and mitigate the impacts of climate change.
• The model can be improved by incorporating additional data sources and features to increase its accuracy and robustness.

References and Datasets

Dataset Source Link : https://www.kaggle.com/datasets/fedesoriano/air-quality-data-in-india