Related Experiment Video
Updated: Jan 15, 2026
09:09
Radio Frequency Identification and Motion-sensitive Video Efficiently Automate Recording of Unrewarded Choice Behavior by Bumblebees
Published on: November 15, 2014
11.3K
Flight and Floral Acoustic Signals for Bee Species Identification
César Augusto Arvelos1, Caique Rocha Resende2, João Pedro Santos Pereira3
1Conservação E Biodiversidade (PPGECB), Programa de Pós-Graduação Em Ecologia, Univ Federal de Uberlândia, Uberlândia, Minas Gerais, Brazil. caarvelos@gmail.com.
Neotropical Entomology
|October 8, 2025
Summary
Automated bee identification using machine learning and buzz sounds is now possible. This study shows high accuracy in distinguishing bee species by analyzing flight and floral sounds.
Area of Science:
- Ecology
- Bioacoustics
- Machine Learning
Background:
- Automated animal identification is crucial for ecological research.
- Tools for automated bee species recognition are underdeveloped.
- Bee identification is vital for biodiversity assessment.
Purpose of the Study:
- To develop and evaluate a machine learning model for identifying five bee species using their buzz sounds.
- To determine the effectiveness of flight and floral buzz sounds for species classification.
- To explore the potential of acoustic monitoring for bee populations.
Main Methods:
- A Random Forest machine learning algorithm was employed.
- Acoustic parameters were extracted from flight and floral buzz sound recordings.
- The fundamental frequency was identified as a key feature for classification.
Main Results:
- Machine learning models achieved 90.94% accuracy with flight buzz and 82.22% with floral buzz.
- Combining both sound types resulted in 95.04% classification accuracy.
- B. pauloensis exhibited lower classification performance due to acoustic overlap.
Conclusions:
- Acoustic features are reliable for species-level bee identification.
- This method offers a non-invasive approach for monitoring bee richness and abundance.
- The findings support the development of automated tools for ecological research and biodiversity assessment.
