University of Twente Student Theses


Deep Unsupervised Representation Learning For Animal Activity Recognition

Voorend, R.W.A. (2021) Deep Unsupervised Representation Learning For Animal Activity Recognition.

This is the latest version of this item.

[img] PDF
Abstract:A generative model as unsupervised representation learning method is indicated to improvethe performance of animal activity recognition (AAR), with intertial measurement unit (IMU)data. Since autoencoders performed well, the introduction of a variational autoencoder (VAE)to the AAR pipeline is tested and investigated. To do this, three feature extraction methodswere built and compared. These were statistical feature extraction, unsupervised representationlearning with a VAE and no feature extraction. Additionally, the size and type of the inputdata were altered to investigate the effects.The results of this research showed that statistical feature extraction performed betterthan representation learning and no feature extraction. This increased the F-score of the AARpipeline with 2%. Representation learning did not improve over using no feature extractionmethod; it scored an F-score that was 3% lower than the pipeline without feature extraction.It was concluded that the addition of feature extraction did improve the classification process,but the classifier that is used might not be compatible with the latent representations forunsupervised representation learning.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:50 technical science in general, 54 computer science
Programme:Creative Technology BSc (50447)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page