An Assessment of Machine Learning Algorithms as Applied to Astronomical Data Sets


Paul Biswas

Document Type


Degree Name

Master of Science (MS)


Physics and Astronomy

Date of Award

Fall 2022


As astronomical datasets increase in scale, the analysis of data becomes less feasible to perform “by hand.” At scales when this becomes an issue, the use of computerized analysis becomes a greater necessity. Modern astronomical surveys are producing terabytes of data available for analysis by the scientific community. Machine learning (ML) algorithms are increasingly being used to classify objects within this data to draw conclusions from the dataset as a whole, but there is a large and growing set of ML algorithms to choose from, and astronomers as a general rule are not familiar with these techniques. This project will explore the efficacy of several the currently available supervised machine learning algorithms to identify and recommend the top choices for researchers more interested in astronomy than in computer science.


Matt A. Wood

Subject Categories

Astrophysics and Astronomy | Computer Sciences | Physical Sciences and Mathematics