Agriculture is the foundation of many economies, including India’s, providing livelihoods, and making a substantial contribution to GDP. On the other hand, selecting crops and projecting yields might be difficult tasks for the younger generation joining the farming industry. Regression and classification algorithms are used in a unique way to forecast yields and suggest crop varieties in order to solve this pressing issue. A detailed analysis of many machine learning algorithms, such as RNN, XGBoost Classifier, KNN, Random Forest, Linear Regression, and Deep Q Network, showed that RNN was more accurate in forecasting ideal harvests. The outcomes demonstrate the effectiveness of the suggested deep learning and machine learning algorithms, offering a comprehensive evaluation using measures including entropy calculation, accuracy, recall, F1 score, sensitivity, and specificity. The suggested method relieves farmers of some of the load by precisely predicting yields for a variety of crops grown in India, giving them the confidence to handle the challenges of agriculture. This creative approach helps control price swings and minimize losses while also enabling the farming community, particularly the younger generation, to make wise decisions that will eventually support the agricultural sector’s long-term expansion.