In the modern agricultural landscape, it’s essential for farmers to adopt advanced agricultural practices to navigate contemporary challenges such as uncontrollable costs due to supply-demand imbalances, water shortages, and climatic uncertainties. Agricultural production faces numerous obstacles, including unpredictable climate changes, inadequate irrigation systems, declining soil fertility, and conventional farming methods. Accurate crop yield forecasting is crucial for enhancing agricultural practices, offering surveillance under diverse climatic conditions, and safeguarding crop yields against various weather-related issues. Machine learning emerges as a potent tool in agriculture, utilized for predicting crop recommendations. This study employs machine learning techniques to recommend a wide range of crops, including rice, maize, jute, cotton, coconut, papaya, orange, apple, muskmelon, watermelon, grapes, mango, banana, pomegranate, lentil, black gram, mung bean, moth beans, pigeon peas, kidney beans, chickpea, and coffee, all influenced by variables such as nitrogen, phosphorus, potassium, rainfall, temperature, humidity, and pH levels. Our proposed method involves the deployment of a Fully Connected Deep Neural Network (FC-DNN) based crop recommendation system, which is then compared against various machine learning models, such as decision tree classifier, logistic regression, gradient boosting classifier, random forest classifier, K-Neighbors classifier, support vector machine classifiers, and Naïve Bayes classification, through performance comparisons evaluating their predictive accuracy.