Classification of GALAXY, QSO, and STAR Based on KNN and PCA

Zhichen Lin

2023

Abstract

As science and technology have advanced over the past few years, numerous astronomical measurement technique projects—like the Sloan Digital Sky Survey (SDSS)—have been built and implemented. Many astronomical data has been collected, including the characteristic data of galaxies, stars, and other celestial bodies. The classification of a large amount of astronomical data requires an efficient algorithm. In this paper, a galaxy (GALAXY), star (STAR), Quasi-Stellar Object (QSO) classification model was constructed using machine learning techniques and the Sloan Digital Sky Survey - DR18 dataset. Different algorithms, including K-Nearest Neighbors (KNN) and Principal Component Analysis (PCA), were used to build this model. The obtained model in this paper exhibits good performance indicators, with accuracy rates of 96%, 98%, 96%, and 98%, respectively. To decrease the dimensionality of the data, the author employed PCA and discovered that certain information in the data was irrelevant to the classification. Discarding these irrelevant features can speed up the training process. The importance of classifying celestial bodies based on astronomical data is evident, as it helps people better understand the composition and evolution of the universe and has significant implications for predicting and explaining astronomical phenomena. However, the same type of celestial body may have significant differences in certain features and practical scenarios, so a more extensive and higher-quality training set is needed to train better-performing models. These models can help people classify celestial bodies more quickly and accurately

Paper Citation

in Harvard Style

Lin Z. (2023). Classification of GALAXY, QSO, and STAR Based on KNN and PCA. In Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML; ISBN 978-989-758-705-4, SciTePress, pages 54-60. DOI: 10.5220/0012814900003885

in Bibtex Style

@conference{daml23,
author={Zhichen Lin},
title={Classification of GALAXY, QSO, and STAR Based on KNN and PCA},
booktitle={Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML},
year={2023},
pages={54-60},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012814900003885},
isbn={978-989-758-705-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Data Analysis and Machine Learning - Volume 1: DAML
TI - Classification of GALAXY, QSO, and STAR Based on KNN and PCA
SN - 978-989-758-705-4
AU - Lin Z.
PY - 2023
SP - 54
EP - 60
DO - 10.5220/0012814900003885
PB - SciTePress