Anomalous Behavior Detection in Galaxies and Exoplanets using ML & DL Techniques

There has been a rapid increase in astronomical data in terms of both size and complexity. To process this extremely large amount of data, machine learning and deep learning methods can be instrumental in arriving at results and further processing relevant data needed by the researchers. In this project, the Inception V3 and Faster RCNN models are employed for determining the morphological type of galaxies along with detection of galaxies in images for the Galaxy Zoo dataset. For detecting the exoplanets, Astronet is used to determine whether the astronomical body is an exoplanet or not on the Kepler time series data. The machine learning method: SGD classifier is predicting the exoplanet habitability, which is tested on PHL’s exoplanet catalogue.