UR: Using Machine Learning Techniques to Analyze Radial Velocity Data to Find Exoplanets

The Undergraduate Research Series is where we present the research that you are Make. If you are an undergraduate student who has participated in a REU or similar astro research project and would like to share this on Astrobites, please see our submission page for more details. We would also like to hear about your more general research experience!

Rohan Nagavardhan

Ward Melville High School.

This post was written by Rohan Nagavardhan, a high school student at Ward Melville High School with a passion for astronomy and computing. He completed this research at SUNY Stony Brook University over the summer under the supervision of Dr Praveen Tripathi. He plans to present these results to Regeneron STS.

In 1995, the radial velocity method found the first extrasolar planet around a Sun-like star in the constellation Pegasus. This method uses the movement of the original star of a planet around a center of mass in order to detect the planet (see figure below) and has been responsible for the detection of a large number of planets. over the past decades. NASA and other groups such as the European Southern Observatory (ESO) and the Sloan Digital Sky Survey (SDSS) amass large catalogs of radial velocity data for stars with instruments such as the High Accuracy Radial Velocity Planet Searcher ( HARPS) and Multi-object APO Radial Large Area Exoplanet Velocity Survey (MARVELS).

We built a machine learning classifier to analyze the radial velocity time series data of a star and predict whether that star has 0, 1, or more orbiting planets. Our radial velocity dataset consisted of data from NASA’s exoplanet archives. Machine learning models are unable to examine time series data (data that measures a certain amount over a certain period of time, i.e. monitoring the value of a stock over 5 years) and form a prediction; therefore, I had to perform a feature extraction. Feature extraction describes each unique radial velocity data set with a specific set of orbital parameters that point to the data for that specific star. Performing this method repeatedly with the radial velocity data yielded a data set that would be used to train a model and test the performance of the model. There are two ways to measure the performance of a machine learning model: the F-measure score and accuracy percentages. The F measure score analyzes the accuracy of the model and its ability to recall information to create a metric that measures model performance. Percent accuracy, which calculates the number of correct predictions out of all predictions made, is not always the best way to measure a model’s performance, was also used to supplement the F measure scores. training, the model received an accuracy percentage of 81.20%; in addition, the model received an F measurement score of 0.81. These results justify the use of machine learning methods on radial velocity time series data in order to identify exoplanets.

The radial speed method for finding exoplanets. Image Credit: Johan Jarnestad / Royal Swedish Academy of Sciences

Astrobite edited by: Haley wahl

Featured Image Credit: Johan Jarnestad / Royal Swedish Academy of Sciences

Source link

Previous How Shine made Upper West Side tech buzz during the pandemic
Next Reviews | Jeff Bezos, CEO of Amazon and Rocketman

No Comment

Leave a reply

Your email address will not be published. Required fields are marked *