Dataset Sources¶
This project uses multiple data sources for AQI prediction.
Below is the list of all the datasets we will be using or have used so far¶
| Source | What It Provides | Format | URL |
|---|---|---|---|
| OpenAQ | Real-time + historical PM2.5, NO2, PM10, O3 for Delhi | JSON API / CSV | openaq.org |
| CPCB (India) | Official govt AQI data by station | CSV download | cpcb.nic.in |
| Kaggle - Delhi AQI | Pre-cleaned historical dataset, good for starting | CSV | kaggle.com/datasets |
| OpenWeatherMap | Temperature, humidity, wind — key features | JSON API (free tier) | openweathermap.org |
| NOAA | Historical weather data for Delhi | CSV | noaa.gov |
📚 Delhi AQI Dataset (Kaggle)¶
Source: 🔗 Delhi NCR Air Quality & Pollution Dataset (2020–2025)
Description:
This dataset contains hourly air quality and environmental data collected across major cities in the Delhi NCR region, designed for pollution analysis, forecasting, and machine learning applications. It captures real-world pollution behavior such as severe winter smog, post-Diwali spikes, and monsoon air quality improvements, making it highly suitable for time-series and environmental modeling tasks.