Data mining a three dimensional beehive content monitoring system

Abstract

Data from a novel capacitive sensor array developed at the University of Manitoba, temperature sensors, humidity sensors, and light sensors placed inside a beehive for a summer is analysed using various time series analysis and data mining techniques. The goal of the analyses are to create features relevant in tracking honey production, detecting swarming, and detecting clustering events. The secondary goal of this thesis is to outline the current state of precision beekeeping in order to understand how this novel capacitive sensor array fits into the greater world of precision beekeeping. This work is a first step in the analytic requirements for the ultimate goal of automating beekeeping using data from sensors like these. Data from this novel capacitive sensor array demonstrated performance unrivalled by other content monitoring methods currently used in precision beekeeping.