Smart Ontology-Based System for Recommending Practices in Melon Cultivation

Smart agriculture systems are models that offer significant potential to improve farming practices by providing advanced tools for land analysis, plant monitoring, weed management, and produce estimation. Many small-scale farmers, particularly in major agricultural regions such as East Java, Indonesia, continue to rely on inefficient traditional methods. Therefore, this research aimed to introduce the Smart Agriculture Ontology System, an innovative framework that incorporates computer vision, deep learning, and semantic web technologies to manage agricultural knowledge effectively. The system integrated traditional observational data with sensor data into a unified knowledge graph, accessible via a query language designed for retrieving and manipulating data stored in the graph. In a case analysis at Puspalebo Orchard, Sidoarjo, East Java, this system provided real-time recommendations for seed selection, soil management, irrigation, pest control, and post-harvest handling. The results from this research showed that the system improved productivity and efficiency by delivering accurate, data-driven recommendations, making it a valuable tool for modern farming. Moreover, the methodology was designed to be generalizable and applicable to various agricultural contexts, allowing it to be a versatile method for different crops and farming conditions. The potential incorporation of external data sources, such as weather information, demonstrated the adaptability of the system for future agricultural management.