PROFIT-AWARE CROP FORECASTING USING MACHINE LEARNING AND MARKET DATA

Abstract

Agriculture remains the backbone of India’s economy, yet farmers often face the dual uncertainty of unpredictable crop yields and fluctuating market prices. While several existing solutions address yield forecasting, they fail to incorporate profitability, leaving farmers vulnerable to income instability. This project proposes a Profit-Aware Crop Forecasting System that integrates machine learning–based yield prediction models (Random Forest, XGBoost, Linear Regression, SVR) with market price forecasting models (ARIMA, Prophet, LSTM). The system estimates profitability by combining predicted yield with market price data, deducting input costs, and recommending the most profitable crop. The goal is to empower farmers with reliable, data-driven decisions for sustainable farming and improved income stability.