Automated Classification of Stellar Spectra using the Sloan Digital Sky Survey

Document Type

Oral Presentation

Campus where you would like to present

Ellensburg

Event Website

https://digitalcommons.cwu.edu/source

Start Date

16-5-2019

End Date

16-5-2019

Abstract

The classification of stellar spectra is a fundamental task in stellar astrophysics. There have been many explorations into the automated classification of stellar spectra but few that involve the Sloan Digital Sky Survey (SDSS). The SDSS is used since it is the most important stellar spectra database available today. Stellar spectra from the SDSS are applied to standard classification methods such as K Nearest Neighbors, Random Forest, and Support Vector Machine. The dimensionality of the stellar spectra is reduced using standard Feature Selection methods such as Chi-Squared and Fisher score and with domain specific astronomical knowledge. These methods are utilized to classify the stellar spectra into the two standard star classification schemes, the Harvard Spectral Classification and the Morgan Keenan Luminosity Classes. A physical phenomenon known as redshift causes machine learning complications when classifying stellar spectra. However, classifiers utilizing redshifted stellar spectra performed with high accuracy. This creates a viable option for automated classification of stellar spectra without having to correct for redshift.

Faculty Mentor(s)

Razvan Andonie

Department/Program

Computer Science

Brice_SOURCE2019.pdf (1940 kB)
Slides for SOURCE 2019 presentation Brice

Additional Files

Brice_SOURCE2019.pdf (1940 kB)
Slides for SOURCE 2019 presentation Brice

Share

COinS
 
May 16th, 12:00 AM May 16th, 12:00 AM

Automated Classification of Stellar Spectra using the Sloan Digital Sky Survey

Ellensburg

The classification of stellar spectra is a fundamental task in stellar astrophysics. There have been many explorations into the automated classification of stellar spectra but few that involve the Sloan Digital Sky Survey (SDSS). The SDSS is used since it is the most important stellar spectra database available today. Stellar spectra from the SDSS are applied to standard classification methods such as K Nearest Neighbors, Random Forest, and Support Vector Machine. The dimensionality of the stellar spectra is reduced using standard Feature Selection methods such as Chi-Squared and Fisher score and with domain specific astronomical knowledge. These methods are utilized to classify the stellar spectra into the two standard star classification schemes, the Harvard Spectral Classification and the Morgan Keenan Luminosity Classes. A physical phenomenon known as redshift causes machine learning complications when classifying stellar spectra. However, classifiers utilizing redshifted stellar spectra performed with high accuracy. This creates a viable option for automated classification of stellar spectra without having to correct for redshift.

https://digitalcommons.cwu.edu/source/2019/Oralpres/175