The discovery of exoplanets is a major focus of contemporary astronomical research. Traditional techniques such as radial velocity, transit methods, gravitational microlensing, direct imaging, polarimetry, and astrometry have historically been used to identify exoplanets. However, handling the massive amounts of data collected by satellites requires efficient and effective methods. This paper introduces a machine learning-based approach combined with Synthetic Minority Oversampling Technique (SMOTE) for classifying exoplanetary systems. The study highlights that while machine learning models achieve high baseline accuracy, their performance can degrade when faced with imbalanced datasets. SMOTE addresses this issue by generating synthetic data points for the minority class, ensuring more balanced model predictions. This research utilized K-Nearest Neighbor (KNN), logistic regression, and Random Forest algorithms, with Random Forest achieving the highest predictive accuracy. Data augmentation techniques were applied solely to the training dataset, while the testing dataset underwent standard preprocessing. The results demonstrated a significant improvement in prediction accuracy through the application of SMOTE. These findings contribute to advancing the identification of exoplanets and bring humanity closer to understanding the potential for life beyond earth.