BeeHealthAI is an intelligent, rule-based prediction system that will monitor the health status of a beehive when using in-hive and environmental sensor data. The system harnesses metaheuristic optimization and association rule mining techniques to provide interpretable, data-driven analysis to support open-ended beekeeping.
- uses OOBO (One-to-One Based Optimizer) for feature selection.
- executes the **Apriori algorithm ** for association rule generation.
- supplies interpretable "if-then" rules to classify hive health as Good or Bad.
- analyzes real-world data from smart beehives (HOBOS dataset).
- facilitates sustainable apiculture by enabling proactive management of hives.
- Python (Pandas, NumPy, MLxtend)
- Jupyter Notebook
- Association Rule Mining (Apriori)
- Metaheuristics (OOBO)
The data used for this project was obtained from the HOBOS project and includes sensor readings on temperature, humidity, weight, and bee flow from hives located throughout Germany.
To provide beekeepers and researchers with an interpretable, lightweight, and accurate AI system that predicts honeybee health and assists with decision-making in time for preventable losses.
MIT License