The Internet of Things (IoT), integrated with machine learning (ML), is revolutionizing smart agriculture by enabling data-driven, precision-based farming practices. Despite extensive individual research on IoT and ML applications, a consolidated framework analyzing their synergistic role in enhancing productivity, sustainability, and food safety remains underexplored. This review identifies this gap and provides a comprehensive analysis of how ML algorithms enhance the intelligence of IoT-based agricultural systems. The study evaluates IoT-ML integration across five key areas: soil management, irrigation, crop monitoring and disease detection, livestock monitoring, and food safety. Combining sensor data with optimized ML models significantly improves accuracy, resource efficiency, and early warning systems. This review not only synthesizes current methodologies and their performance metrics but also highlights challenges such as data privacy, energy efficiency, and system scalability. Future research directions include the development of robust, interoperable IoT frameworks, energy-harvesting sensor networks, and adaptive ML models for real-time, large-scale deployment in precision agriculture.