SMART-FARM-IDS: A MACHINE LEARNING APPROACH FOR IOT-BASED AGRICULTURE

Abstract

Smart agriculture leverages Internet of Things (IoT) technologies to enable real-time monitoring of soil, water, and environmental conditions for optimized crop management. This study presents an Arduino-based Smart-Farm Intrusion Detection System (Smart-Farm-IDS) integrating IoT sensors, automation, and machine learning to enhance both productivity and cybersecurity in agriculture. Using models such as Random Forest, KNN, Support Vector Machine, Logistic Regression, Decision Tree, the system detects anomalies and cyber intrusions with over 99% accuracy. The proposed framework demonstrates that combining IoT-driven automation with intelligent intrusion detection can ensure sustainable, efficient, and secure farming operations.