Machine learning for weather-specific crop recommendation

Abstract

Agriculture and its related sectors are unquestionably the most important sources of income in rural India. Additionally, having a big impact on the nation's GDP is the agriculture sector. The sector of agriculture is so large, which is great for the nation. The crop production per hectare, however, falls short of international standards. This is one of the most likely causes of the greater rate of suicide among marginal farmers in India. This research proposed best recommendation system. The proposed system recommends crops for farmers to grow based on input from the farmer’s field, such as the temperature, soil, moisture, and nutrient like NPK, pH, and rainfall. Machine learning algorithms allow for optimal crop selection to be made in light of all relevant parameters. three popular machine learning algorithms were tested in this study which includes the Decision Tree, the Random Forest (RF), and the Logistic Regression. The Random Forest among them demonstrated the highest outcomes with 99.32% accuracy.