This project presents a multi-method AI pipeline for prioritizing exoplanet candidates from the NASA Kepler KOI dataset. Combining from-scratch implementations of Decision Trees, Naive Bayes, K-Means clustering, and Bayesian reasoning with a CNN baseline, the system ranks 1,979 unresolved signals by their likelihood of being genuine exoplanets.
-
Updated
May 6, 2026 - Python