A Machine Learning-Driven Soil Nutrient and Crop Yield Recommendation Platform with Pesticide Suggestions

Crop selection plays a critical role in agriculture. This work is a machine learning-based platform that recommends crops based on soil nutrient levels, predicts crop yield and water content, and suggests appropriate pesticide use. The system takes input data such as soil pH, nitrogen, phosphorus, and potassium levels, and environmental factors like temperature and rainfall to predict the optimal crop for a specific region. Additionally, it provides pesticide suggestions based on predicted crop and current pest situations. The input data is collected from soil nutrient analysis, weather data, and crop data from internet sources. The system is developed from real-world data from the Indian subcontinent and assessed using machine learning methods such as Logistic Regression, Random Forest, and Support Vector. Results demonstrate the potential to improve crop yields and promote sustainable learning practices. The system is validated using evaluating parameter metrics, accuracy, and it is proved that the Random Forest is capable of providing accurate predictions.