Unsupervised source separation for multi-label classification

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

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 proceeding › Conference 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 language	English
Title of host publication	30th European Signal Processing Conference (EUSIPCO 2022): Proceedings
Publisher	IEEE
Pages	1686-1690
Number of pages	5
ISBN (Electronic)	9789082797091
ISBN (Print)	9781665467995
Publication status	Published - 18 Oct 2022
Event	30th European Signal Processing Conference - Crowne Plaza, Belgrade, Serbia Duration: 29 Aug 2022 → 2 Sep 2022 https://2022.eusipco.org/ (Link to conference website)

Publication series

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

Conference

Conference	30th European Signal Processing Conference
Abbreviated title	EUSIPCO 2022
Country/Territory	Serbia
City	Belgrade
Period	29/08/22 → 2/09/22
Internet address	https://2022.eusipco.org/ (Link to conference website)

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

Documents and Links

Mitiche, I. et al (2022) Unsupervised source separation for multi-label classification
Copyright © 2022 IEEE. This is a post-peer review, pre-copy edited version of the article and may differ slightly from the final published version of the work. This work is made available in line with the publisher policy. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.
Author accepted manuscript, 2.59 MB

Cite this

@inproceedings{059167b8cae348c69f4c8278dfe36ad9,

title = "Unsupervised source separation for multi-label classification",

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.",

keywords = "Multi-label classification, source separation, variation auto-encoder, EMI technique, high-voltage asset diagnosis",

author = "Imene Mitiche and Alireza Salimy and Philip Boreham and Tony McGrail and Alan Nesbitt and Gordon Morison",

year = "2022",

month = oct,

day = "18",

language = "English",

isbn = "9781665467995",

publisher = "IEEE",

pages = "1686--1690",

booktitle = "30th European Signal Processing Conference (EUSIPCO 2022): Proceedings",

note = "30th European Signal Processing Conference, EUSIPCO 2022 ; Conference date: 29-08-2022 Through 02-09-2022",

url = "https://2022.eusipco.org/",

}

TY - GEN

T1 - Unsupervised source separation for multi-label classification

AU - Mitiche, Imene

AU - Salimy, Alireza

AU - Boreham, Philip

AU - McGrail, Tony

AU - Nesbitt, Alan

AU - Morison, Gordon

PY - 2022/10/18

Y1 - 2022/10/18

N2 - 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.

AB - 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.

KW - Multi-label classification

KW - source separation

KW - variation auto-encoder

KW - EMI technique

KW - high-voltage asset diagnosis

UR - https://ieeexplore.ieee.org/document/9909605/

M3 - Conference contribution

SN - 9781665467995

SP - 1686

EP - 1690

BT - 30th European Signal Processing Conference (EUSIPCO 2022): Proceedings

PB - IEEE

T2 - 30th European Signal Processing Conference

Y2 - 29 August 2022 through 2 September 2022

ER -

Unsupervised source separation for multi-label classification

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Documents and Links

Other files and links

Fingerprint

Cite this