The challenges of sustainable agriculture include ensuring optimal crop selection amidst unpredictable climatic changes, market dynamics, and limited natural resources. This research introduces a machine learning-based approach for intelligent crop selection, which leverages data-driven insights to optimize agricultural productivity. By analyzing soil properties, climate data, and market trends, the system provides farmers with actionable recommendations to maximize yields and profitability. Machine learning models such as Random Forest and Artificial Neural Networks (ANN) were employed, achieving a predictive accuracy of 95% during testing. This system demonstrates significant potential to transform traditional farming practices into smart, efficient agricultural decision-making processes.