Artificial intelligence has advanced significantly in various fields, including physics, where machine learning and deep learning techniques, such as boosted decision trees and neural networks, have been widely applied. In high-energy physics, these methods have improved tasks like image recognition and data analysis. This study focuses on developing machine learning algorithms for the classification of electromagnetic and hadronic showers in space calorimeter experiments. Using Monte Carlo simulations and feature engineering, the results highlight the potential of AI to enhance classification accuracy in space calorimetry.