AGROSMART: A MACHINE LEARNING DRIVEN SYSTEM FOR PREDICTIVE AGRICULTURE

Computer Science Engineering

Agriculture increasingly depends on data-driven methods to improve productivity and resource utilization. This paper presents AgroSmart, a machine learning-based decision support system developed to assist farmers in crop selection, fertilizer planning, and crop yield prediction. The proposed system analyzes soil and environmental parameters such as nitrogen, phosphorus, potassium content, soil pH, rainfall, temperature, and humidity. For crop recommendation, a Random Forest classifier is used to identify the most suitable crop based on input conditions. For yield prediction, a Linear Regression model estimates expected production from historical patterns and current parameters. The system is implemented in Python using a Flask backend with a user-friendly web interface that allows users to enter field data and receive recommendations in real time. The proposed platform can help reduce inefficient fertilizer use, support informed farm decisions, and improve planning at the field level. AgroSmart demonstrates how machine learning can be applied in practical farming environments to support more efficient and sustainable agriculture.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.