Integrated Smart Crop Management System Using Explainable Artificial Intelligence (XAI)
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Farmers detect crop stress problems, which arise from water shortages and extreme heat and nutrient deficiencies, after visible plant damage has already occurred, which leads to substantial crop yield declines. The current digital agriculture systems only provide disease detection capabilities together with methods for analyzing events after they occur, while they do not offer systems for monitoring agricultural conditions or simple tools that help users make decisions. The research develops an AI/ML-based system to support decision-making through its capability to detect crop stress at an early stage by processing multiple data types which include structured soil parameters and short-term weather forecasts and unstructured farmer field notes processed using TextBlob. The system produces stress forecasts for the next 7 to 10 days through its ensemble models which include Random Forest and XGBoost and their optimization system that uses grid search and cross-validation techniques. The RESTful backend system facilitates data handling together with model execution and Explainable AI(XAI) integration because SHAP-based model interpretations become available through a web interface. The system allows resource-limited users to enter data through both manual and document-based methods, while it shows accurate results with interpretable predictions on three benchmark datasets and synthetic data.
Fields of Study & Subjects
Smart Agriculture and AI
Artificial Intelligence in Healthcare
Remote Sensing in Agriculture
Publication details
- Journal / source
- Zenodo (CERN European Organization for Nuclear Research)
- Publication date
- 27 March 2026
- Language
- EN
- Document type
- Preprint