UrbanHeat AI

▪︎ Inspiration

Rapid urbanization and declining green cover are increasing heat levels in cities. UrbanHeat AI was developed to identify heat hotspots and support sustainable urban planning.

▪︎What it does

UrbanHeat AI predicts street-level heat conditions using environmental factors such as humidity and green cover. It classifies areas into heat zones (Hot, Moderate, Cool), visualizes them on an interactive map, and provides actionable climate recommendations.

▪︎ How we built it

The solution was developed using a Flask-based web application, with a Linear Regression model for prediction. Pandas was used for data processing, and Folium enabled interactive geospatial visualization.

▪︎Challenges

Key challenges included data preprocessing, defining accurate classification thresholds, and integrating machine learning outputs with real-time map visualization.

▪︎Accomplishments

We successfully built an end-to-end AI-powered system that combines predictive modeling with geospatial insights to deliver practical, real-world solutions.

▪︎What we learned

This project strengthened our understanding of machine learning deployment, web development with Flask, and the importance of data quality in predictive systems.

▪︎What’s next

Future enhancements include integrating real-time weather data, improving model accuracy with advanced algorithms, and expanding the system into a mobile platform.

Built With

• application
• built-with-languages:-python
• css-frameworks-&-libraries:-flask
• folium
• html
• jupyter
• leaflet.js
• machine
• pandas
• regression
• scikit-learn
• web