Summary
This paper presents an IoT-enabled machine learning framework that integrates soil and environmental sensors with edge computing and cloud management for adaptive irrigation scheduling. The system employs a Random Forest classifier to make real-time irrigation decisions, achieving 97.8% prediction accuracy and substantially reducing water consumption relative to conventional threshold-based approaches. The framework is proposed as a lightweight, cost-effective solution for precision agriculture and climate adaptation.
Regional applicability
The framework's applicability to United Kingdom conditions would depend on local water scarcity priorities, sensor deployment costs, and compatibility with existing UK precision agriculture infrastructure. The climate-adaptation focus and water-efficiency gains are relevant to UK farming under increasing rainfall variability, though the paper does not report UK-specific validation or compare performance under UK soil and weather conditions.
Key measures
Model accuracy (97.8% for Random Forest classifier), water consumption reduction (31.7% vs. traditional irrigation), comparison of classifier performance (Random Forest vs. Decision Tree vs. SVM)
Outcomes reported
The study developed and evaluated an IoT-enabled machine learning system for smart irrigation, measuring model accuracy in irrigation prediction and water consumption reduction compared to conventional practices. The Random Forest classifier achieved 97.8% accuracy and the framework reduced water consumption by 31.7% versus traditional irrigation methods.
Topic tags
Dig deeper with Pulse AI.
Ask about this record, its theme or its relevance to UK farming and policy. Pulse AI uses selected catalogue evidence and cites the sources it draws on.