International Journal of Scientific Research and Engineering Development

This paper presents an IOT-BASED PRECISION AGRICULTURE SYSTEM USING MACHINE LEARNING ALGORITHMS designed to provide real-time, localized agricultural advice to farmers. The system leverages an ESP32 microcontroller as the edge device to collect critical environmental data from rainfall and DHT11 (temperature and humidity) sensors. The data is processed locally on the device using a pre-trained Random Forest Classifier model, which was trained on a comprehensive Kaggle dataset. The system features a push button for user interaction and displays the recommended crop on an LCD I2C screen. This approach minimizes latency and dependence on a constant internet connection, making it ideal for rural farming environments. The use of the Everywhere ML framework facilitates the deployment of the machine learning model onto the constrained hardware. The results demonstrate the system's ability to provide accurate and timely recommendations, empowering farmers to make data-driven decisions for improved crop yield.