Deep Learning for Galaxy Morphology Classification in Large-Scale Surveys

Parkar, Omkar Saurabh (2024) Deep Learning for Galaxy Morphology Classification in Large-Scale Surveys. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (922kB) | Preview
Preview	PDF (Configuration Manual) Download (2MB) | Preview

Abstract

In this thesis, the deep learning techniques and their applications are explored on the classification of the galaxy morphologies in the large-scale surveys of astronomical data. The dataset used in this study is derived from the Sloan Digital Sky Survey (SDSS) which is the combination of photometric and spectroscopic data across five spectral bands. This research thesis uses a robust framework for automating the galaxy classification and the dataset uses 500,000 records with comprehensive attribute listing. These attributes include the astronomical coordinates, photometric measurements, observational details, and class labels (galaxy, star, quasar). The methodology involved in this research report includes a detailed data preparation, data cleaning, normalization, and transformation. This is done so that a better and optimized model can be trained and developed. The evaluation for such models used in this report is conducted by the traditional evaluation metrics like accuracy, precision, recall, F1-score, and confusion matrices. Resultant models are comprehensive and show the accuracy of classifying relevant galaxies into their respective morphological categories. This report also shows the key findings to be data augmentation and class imbalance, which are addressed to achieve even better classification accuracies across different galaxy types

Item Type:	Thesis (Masters)
Supervisors:	Name Email Kelly, John UNSPECIFIED
Uncontrolled Keywords:	Galaxy morphology; SDSS; Deep learning; Data analysis; Classification; EDA; Data balancing
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:	25 Aug 2025 09:17
Last Modified:	25 Aug 2025 09:17
URI:	https://norma.ncirl.ie/id/eprint/8607

Actions (login required)

View Item