Unsupervised source separation for multi-label classification

Imene Mitiche, Alireza Salimy, Philip Boreham, Tony McGrail, Alan Nesbitt, Gordon Morison

Research output: Chapter in Book/Report/Conference proceedingConference contribution

15 Downloads (Pure)

Abstract

This paper exploits blind source separation for the purpose of multi-label classification (MLC). The proposed method addresses the multi-label classification problem in two stages, source separation of data consisting of two or more observed classes followed by a single-label classification. This method is evaluated on two data-sets: the new application to electromagnetic interference (EMI) signals that are used for high-voltage power asset condition monitoring, and overlapping handwritten digits (multi-MNIST) for further model validation. Our approach is compared to other deep neural networks such as variational auto-encoder (VAE) and capsule network (CapsNet). Results show high average precision on both data-sets and outperform in most metrics other deep auto-encoding networks that infer label decision from the latent variables distribution.
Original languageEnglish
Title of host publication30th European Signal Processing Conference (EUSIPCO 2022): Proceedings
PublisherIEEE
Pages1686-1690
Number of pages5
ISBN (Electronic)9789082797091
ISBN (Print)9781665467995
Publication statusPublished - 18 Oct 2022
Event30th European Signal Processing Conference - Crowne Plaza, Belgrade, Serbia
Duration: 29 Aug 20222 Sep 2022
https://2022.eusipco.org/ (Link to conference website)

Publication series

Name
ISSN (Print)2219-5491
ISSN (Electronic)2076-1465

Conference

Conference30th European Signal Processing Conference
Abbreviated titleEUSIPCO 2022
Country/TerritorySerbia
CityBelgrade
Period29/08/222/09/22
Internet address

Keywords

  • Multi-label classification
  • source separation
  • variation auto-encoder
  • EMI technique
  • high-voltage asset diagnosis

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Unsupervised source separation for multi-label classification'. Together they form a unique fingerprint.

Cite this