Original title: MSPB: a longitudinal multi-sensor dataset with phenotypic trait measurements from honey bees
Authors: Yi Zhu, Mahsa Abdollahi, Ségolène Maucourt, Nico Coallier, Heitor R. Guimarães, Pierre Giovenazzo, Tiago H. Falk
The article unveils a comprehensive dataset from honey bee colonies collected over a year in Québec, Canada. This unique dataset, gathered from 53 hives, integrates various sensor data like audio features, temperature, and humidity, alongside detailed phenotypic measurements. These measurements cover hive population, brood cells, Varroa destructor infestation, behavioral traits, honey production, and winter mortality. Notably, experts in apicultural science annotated these diverse measurements, enabling broader analyses. The article outlines the data collection process, pre-processing steps, and its content. It showcases the distribution of phenotypic data and patterns in sensor data. Additionally, it demonstrates practical applications like predicting winter mortality, estimating hive population, and detecting an active queen bee using machine learning and sensor data analysis. This dataset promises valuable insights for studying honey bee health and behavior, offering numerous possibilities for further exploration and understanding.
Original article: https://arxiv.org/abs/2311.10876