A Novel Methodology for Hunting Exoplanets in Space Using Machine Learning

Abstract

INTRODUCTION: Exoplanet exploration outside of our solar system has recently attracted attention among astronomers worldwide. The accuracy of the currently used detection techniques, such as the transit and radial velocity approaches is constrained. Researchers have suggested utilizing machine learning techniques to create a prediction model to increase the identification of exoplanets beyond our milky way galaxy.
 OBJECTIVES: The novel method proposed in this research paper builds a prediction model using a dataset of known exoplanets and their characteristics, such as size, distance from the parent star, and orbital period. The model is then trained using this data based on machine learning methods that Support Vector Machines and Random Forests.
 METHODS: A different dataset of recognized exoplanets is used to assess the model’s accuracy, and the findings are compared with in comparison to accuracy rates of the transit and radial velocity approaches. 
 RESULTS: The prediction model created in this work successfully predicts the presence of exoplanets in the test data-set with an accuracy rate of over 90 percent.
 CONCLUSION: This discovery shows the promise and confidence of machine learning techniques for exoplanet detection.