Prediction of Crop Recommendation Technique Using Supervised Machine Learning Method

This paper presents the implementation of a machine learning-based crop recommendation system using soil composition data. Through extensive exploratory data analysis (EDA), key soil parameters-such as temperature, humidity, nitrogen, phosphorus, and potassium levels-are identified as crucial for optimal crop growth. The dataset is thoroughly analyzed to determine crop-specific requirements, and several machine learning models, including K-Nearest Neighbors, Decision Tree, and Random Forest, are trained and evaluated. Among these models, the Random Forest Classifier achieved the highest accuracy, with 99.59%, in predicting the most suitable crop for given soil conditions. For instance, the model recommends approximately 70mm of rainfall for optimal rice growth. The results demonstrate a robust model capable of generalizing well across various soil compositions, offering a valuable tool for precision agriculture. This predictive system provides farmers with a reliable and efficient method to optimize crop yield, supporting sustainable farming practices. The study successfully meets its objectives, with minimal deviations, and sets the stage for future advancements in AI-driven smart agriculture solutions.