LAUSR

Exoplanet detection is an extremely active research topic in astronomy. Researchers in the past have attempted to detect exoplanets using conventional methods like Radial Velocity, Transit Method, Gravitational Microlensing, Direct Imaging, Polarimetry, Astrometry, etc. While the approaches undertaken for all these studies vary, many of the research works conducted are based on the change in flux (light intensity). Based on the same characteristic, we explore yet another method of detecting exoplanets in space, using Artificial Intelligence. We rely on several machine learning algorithms like Decision Trees, Support Vector Machines, Logistic Regression, Random Forest Classifier, Multilayer Perceptron (MLP), Convolutional Neural Networks (CNN), as baseline algorithms and introduce an Ensemble-CNN model to draw out comparisons between the different machine learning models. The performance of the models has been evaluated using parameters like Accuracy, Precision, Sensitivity, and Specificity. Our results denote that the proposed Ensemble-CNN model performs relatively better for detecting exoplanets with an accuracy of 99.62%. The research will be useful in the fields of Astronomy as well as Artificial Intelligence and would be of substantial importance to physicists, cosmologists, scientists, researchers, academicians, industry experts, and machine learning experts who work in areas related to (or closely related to) exoplanet detection.