We are Team Kepler-452b, participants in the NASA Space Apps Challenge 2025. Our challenge was: “A World Away: Hunting for Exoplanets with AI.”

To address this, we developed ExoDetect, an AI-powered web application that automatically identifies exoplanets from NASA’s open-source datasets (Kepler, K2, and TESS). Using the transit method, our model analyzes light curves and orbital parameters—such as orbital period, transit duration, and planetary radius—to classify celestial objects as confirmed exoplanets, candidates, or false positives. Built with advanced machine learning techniques like XGBoost and ensemble pipelines, ExoDetect achieves high accuracy through optimized preprocessing and feature selection.

The interactive Streamlit interface allows scientists and enthusiasts to upload or manually input data for instant classification, automating the detection process and accelerating discoveries. Our project bridges the gap between AI and space exploration, making astronomical research more accessible to everyone.

For more info about our project visit: Kepler-452b