Yassine854/AQI-Prediction-for-Tunisia

# Tunisia Air Quality Predictor A web-based tool to predict the Air Quality Index (AQI) across Tunisia using a machine learning regression model trained on environmental and meteorological data. --- ## Overview Tunisia Air Quality Predictor is a React application that allows users to estimate air quality based on input parameters such as particulate matter concentration (PM1), temperature, humidity, and time of day. The app connects to a regression-based machine learning model hosted on Hugging Face Spaces, which outputs a predicted AQI value and corresponding air quality category. --- ## Features - Predicts Tunisia’s Air Quality Index (AQI) using environmental factors. - Interactive and responsive interface built with React and Tailwind CSS. - Provides visual AQI category feedback (Good, Moderate, Unhealthy, etc.). - Includes dynamic gradients and smooth animations for better user experience. - Input validation and error handling for missing fields or connection issues. --- ## The Machine Learning Model The model behind this application is a regression-based predictive model trained on Tunisian air quality and weather data. It estimates AQI from several input variables: | Feature | Description | |----------|-------------| | pm1 | Particulate matter (PM1) concentration (µg/m³) | | relativehumidity | Relative humidity (%) | | temperature | Ambient temperature (°C) | | hour | Hour of the day (0–23) | | month | Month of the year (1–12) | | dayofweek | Day of the week (0–6) | **Example API Response:** ```json { "predicted_aqi": 42.6, "category": "Moderate" }