Abstract

Artificial intelligence has had a significantly smaller impact on space exploration than on other domains. This research analyzes regression models used to predict the Earth Similarity Index (ESI). ESI is an extremely essential statistic for determining exoplanetary habitability. Recognizing the parallels between Earth and exoplanets is critical in determining the habitability of these planets. To capture the complex connections found in ESI data, this study objectively compares the efficiency of various regression models, including Linear Regression, Lasso Regression, Random Forest Regression, Support Vector Regression (SVR), and Huber Regression. These models are utilized to predict ESI, and their performance on this data is assessed to provide a clear picture of AI in predicting ESI. This paper addresses ESI prediction using AI as the initial step in predicting whether exoplanets can potentially support life using AI.