Research

Paper

TESTING March 12, 2026

Exploring the Viability of Fisher Discriminants in Galaxy Morphology Classification

Authors

Sazatul Nadhilah Zakaria, Santtosh Muniyandy, John Y. H. Soo

Abstract

One of the major challenges in astronomy involves accurately classifying galaxies, particularly distinguishing between different galaxy types. While many complex algorithms have shown strong performance in classification tasks, their complexity often results in longer processing times and increased difficulty in understanding. This study addresses this issue by exploring the viability of Fisher discriminants, a much simpler algorithm, in performing galaxy morphology classification. We tested four machine learning algorithms: the Fisher discriminant, Artificial Neural Networks (ANNs), Boosted Decision Trees (BDTs), and k-Nearest Neighbours (kNNs) to classify galaxies by the shape of their central bulges. Using data from the Sloan Digital Sky Survey (SDSS), we utilised five pre-processing transformations: normalisation, decorrelation, principal component analysis (PCA), uniformisation, and Gaussianisation, and classified the shape of central bulge into either rounded or no-bulge, based on the Galaxy Zoo Decision Tree. When compared to the Galaxy Zoo 2 (GZ2) labels, the Fisher discriminant with uniformisation obtained the highest accuracy score of 0.9310, outperforming ANN, BDT, and kNN by 1.93%, 0.42%, and 3.08%, respectively.

Metadata

arXiv ID: 2603.11652
Provider: ARXIV
Primary Category: astro-ph.IM
Published: 2026-03-12
Fetched: 2026-03-13 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.11652v1</id>\n    <title>Exploring the Viability of Fisher Discriminants in Galaxy Morphology Classification</title>\n    <updated>2026-03-12T08:22:00Z</updated>\n    <link href='https://arxiv.org/abs/2603.11652v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.11652v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>One of the major challenges in astronomy involves accurately classifying galaxies, particularly distinguishing between different galaxy types. While many complex algorithms have shown strong performance in classification tasks, their complexity often results in longer processing times and increased difficulty in understanding. This study addresses this issue by exploring the viability of Fisher discriminants, a much simpler algorithm, in performing galaxy morphology classification. We tested four machine learning algorithms: the Fisher discriminant, Artificial Neural Networks (ANNs), Boosted Decision Trees (BDTs), and k-Nearest Neighbours (kNNs) to classify galaxies by the shape of their central bulges. Using data from the Sloan Digital Sky Survey (SDSS), we utilised five pre-processing transformations: normalisation, decorrelation, principal component analysis (PCA), uniformisation, and Gaussianisation, and classified the shape of central bulge into either rounded or no-bulge, based on the Galaxy Zoo Decision Tree. When compared to the Galaxy Zoo 2 (GZ2) labels, the Fisher discriminant with uniformisation obtained the highest accuracy score of 0.9310, outperforming ANN, BDT, and kNN by 1.93%, 0.42%, and 3.08%, respectively.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='astro-ph.IM'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='astro-ph.CO'/>\n    <published>2026-03-12T08:22:00Z</published>\n    <arxiv:comment>19 pages, 5 figures, this is the English translation of the paper published in Malay</arxiv:comment>\n    <arxiv:primary_category term='astro-ph.IM'/>\n    <arxiv:journal_ref>Sains Malaysiana 54(7) (2025) 1785-1796</arxiv:journal_ref>\n    <author>\n      <name>Sazatul Nadhilah Zakaria</name>\n    </author>\n    <author>\n      <name>Santtosh Muniyandy</name>\n    </author>\n    <author>\n      <name>John Y. H. Soo</name>\n    </author>\n    <arxiv:doi>10.17576/jsm-2025-5407-12</arxiv:doi>\n    <link href='https://doi.org/10.17576/jsm-2025-5407-12' rel='related' title='doi'/>\n  </entry>"
}