Ensemble Machine learning to Detect Exoplanets

Petkar, Visha (2024) Ensemble Machine learning to Detect Exoplanets. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (600kB) | Preview

Abstract

In the last 6 decades the exploration of space has unveiled some of the most profound mysteries of our universe, revealing multitudes of celestial objects and wonders, which includes distant exoplanetary systems. Leveraging on the vast data that was obtained by the earlier Kepler and K2 missions, this report aims to present an approach to detect exoplanets by analysing the fluctuations of light curves with a combination of data from the Kepler, K2 and TESS missions. Four models will be trained – Convolutional Neural Network (CNN) utilizing GPU acceleration, Support Vector Machine (SVM), K-Nearest neighbour and Random Forest to identify the subtle signatures of exoplanetary transits within the fluctuations of light. Post the models achieving a satisfactory performance, an ensemble script combining 3 models was used to evaluate its performance in identifying exoplanets from the light curve data obtained from all 3 sources with test data that the models had never analyzed. The results of this research showcases that an ensemble model with CNN, K-NN and Random forest achieved an accuracy of 0.62, precision of 0.66, recall(sensitivity) of 0.70, specificity of 0.51 and an F1 score of 0.68. This indicates that the ensemble approach, particularly leveraging KNN, exhibits promising performance in accurately identifying exoplanets from the analyzed light curve data, thus contributing significantly to the field of exoplanetary research.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Muntean, Christina Hava UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QB Astronomy Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	05 Jun 2025 15:14
Last Modified:	05 Jun 2025 15:16
URI:	https://norma.ncirl.ie/id/eprint/7769

Actions (login required)

View Item