Applied Intelligence and Computing
Stellar Classification using Linear Regression: Insights into Predictive Model Optimization
Authors: Arpanpreet Kaur, Kanwarpartap Singh Gill, Sonal Malhotra and Swati Devliyal
Publishing Date: 05-11-2024
ISBN: 978-81-955020-9-7
Abstract
A thorough analysis of star classification using linear regression is summarised in the abstract of this research report. In particular, the study looks at how well different characteristics may predict star type, such as absolute magnitude, colour, relative luminosity, relative radius, absolute temperature, and spectral class. A wide variety of stars, from Red Dwarfs to HyperGiants, are included in the collection, which gives a fertile ground for investigation. The effectiveness of this strategy in stellar classification is highlighted by the linear regression model's 90% accuracy in predicting star kinds. The model's performance is assessed across various star types using precision, recall, and F1-score measures. This allows us to understand its strengths and limits. The abstract provides context for the publication by outlining its main points, highlighting the research's relevance to our knowledge of stars and astronomy as a whole, and summarising its main results.
Keywords
Deep Learning, Plant Disease Detection, Convolutional Neural Networks, Occlusion Experiment, Saliency Map, PyTorch, Model Visualization, Interpretability, Agriculture.
Cite as
Arpanpreet Kaur, Kanwarpartap Singh Gill, Sonal Malhotra and Swati Devliyal, "Stellar Classification using Linear Regression: Insights into Predictive Model Optimization", In: Mukesh Saraswat and Rajani Kumari (eds), Applied Intelligence and Computing, SCRS, India, 2024, pp. 49-55. https://doi.org/10.56155/978-81-955020-9-7-6