Honeybee Colony Growth Period Recognition Based on Multivariate Temperature Feature Extraction and Machine Learning

Area of Science:

• Agricultural Science
• Data Science
• Entomology

Background:

• Accurate identification of bee colony growth periods is crucial for effective beekeeping management.
• Traditional methods rely on manual experience, which can be subjective and time-consuming.

Purpose of the Study:

• To develop an intelligent system for recognizing bee colony growth periods.
• To leverage multivariate temperature data and machine learning for enhanced accuracy and efficiency.

Main Methods:

• Collected year-round temperature data from 38 bee hives across two regions.
• Extracted 17 time-domain characteristic indices and analyzed feature sensitivity across different time scales.
• Applied Principal Component Analysis (PCA) for dimensionality reduction and feature enhancement.
• Utilized six machine learning algorithms (supervised and unsupervised) for growth period identification.

Main Results:

• The extracted temperature features effectively characterize bee colony growth periods.
• The Backpropagation (BP) neural network achieved the best performance, with a Mean Absolute Error (MAE) of 1.45%.
• The method demonstrated practicability across different geographical regions.

Conclusions:

• The proposed approach offers a data-driven and intelligent solution for identifying bee colony growth periods.
• This technique can significantly improve beekeeping decision-making and promote colony development.
• The findings highlight the potential of machine learning in precision apiculture.