AI-Assisted Farming for Crop Recommendation & Virtual Assistance Application

Abstract

It is a fact that agriculture in India contributes significantly to employment and economic growth. The common problem Indian farmers experience is choosing the right crop for their soil, which reduces productivity. The farmers can resolve this problem through precision agriculture. Precision agriculture is characterized by a soil database collected on the farm, crop information provided by agricultural experts, and soil parameters from soil testing labs. As soon as the soil testing lab provides data to the recommendation system, the system will utilize the information and create an ensemble model. For the recommendation, we need a few parameters like NPK (nitrogen, phosphorus, potassium) ratios, and PH values. This project gives full assistance to the farmers virtually. By doing so we can prevent the suicide rates of farmers across the country. This paper concentrates specifically about the prediction of crops. The collected data about the plants and their NPK values are fed into ML models. The preferred model for crop prediction is Random Forest.