Enhancing supervised classifications with metamorphic relations

We report on a novel use of metamorphic relations (MRs) in machine learning: instead of conducting metamorphic testing, we use MRs for the augmentation of the machine learning algorithms themselves. In particular, we report on how MRs can enable enhancements to an image classification problem of ima...

Full description

Bibliographic Details
Main Authors: Xu, Liming, Towey, Dave, French, Andrew P., Benford, Steve, Zhou, Zhi Quan, Chen, Tsong Yueh
Format: Conference or Workshop Item
Language:English
English
Published: 2018
Online Access:https://eprints.nottingham.ac.uk/53764/
_version_ 1848798986300817408
author Xu, Liming
Towey, Dave
French, Andrew P.
Benford, Steve
Zhou, Zhi Quan
Chen, Tsong Yueh
author_facet Xu, Liming
Towey, Dave
French, Andrew P.
Benford, Steve
Zhou, Zhi Quan
Chen, Tsong Yueh
author_sort Xu, Liming
building Nottingham Research Data Repository
collection Online Access
description We report on a novel use of metamorphic relations (MRs) in machine learning: instead of conducting metamorphic testing, we use MRs for the augmentation of the machine learning algorithms themselves. In particular, we report on how MRs can enable enhancements to an image classification problem of images containing hidden visual markers ("Artcodes"). Working on an original classifier, and using the characteristics of two different categories of images, two MRs, based on separation and occlusion, were used to improve the performance of the classifier. Our experimental results show that the MR-augmented classifier achieves better performance than the original classifier, algorithms, and extending the use of MRs beyond the context of software testing.
first_indexed 2025-11-14T20:28:29Z
format Conference or Workshop Item
id nottingham-53764
institution University of Nottingham Malaysia Campus
institution_category Local University
language English
English
last_indexed 2025-11-14T20:28:29Z
publishDate 2018
recordtype eprints
repository_type Digital Repository
spelling nottingham-537642018-09-06T09:59:15Z https://eprints.nottingham.ac.uk/53764/ Enhancing supervised classifications with metamorphic relations Xu, Liming Towey, Dave French, Andrew P. Benford, Steve Zhou, Zhi Quan Chen, Tsong Yueh We report on a novel use of metamorphic relations (MRs) in machine learning: instead of conducting metamorphic testing, we use MRs for the augmentation of the machine learning algorithms themselves. In particular, we report on how MRs can enable enhancements to an image classification problem of images containing hidden visual markers ("Artcodes"). Working on an original classifier, and using the characteristics of two different categories of images, two MRs, based on separation and occlusion, were used to improve the performance of the classifier. Our experimental results show that the MR-augmented classifier achieves better performance than the original classifier, algorithms, and extending the use of MRs beyond the context of software testing. 2018-05-27 Conference or Workshop Item PeerReviewed application/pdf en https://eprints.nottingham.ac.uk/53764/1/MRAugmentedClassifierMET2018.pdf application/pdf en https://eprints.nottingham.ac.uk/53764/2/p46-xu.pdf Xu, Liming, Towey, Dave, French, Andrew P., Benford, Steve, Zhou, Zhi Quan and Chen, Tsong Yueh (2018) Enhancing supervised classifications with metamorphic relations. In: 3rd International Workshop on Metamorphic Testing - MET '18, 27 May 2018, Gothenburg, Sweden. https://doi.org/10.1145/3193977.3193978
spellingShingle Xu, Liming
Towey, Dave
French, Andrew P.
Benford, Steve
Zhou, Zhi Quan
Chen, Tsong Yueh
Enhancing supervised classifications with metamorphic relations
title Enhancing supervised classifications with metamorphic relations
title_full Enhancing supervised classifications with metamorphic relations
title_fullStr Enhancing supervised classifications with metamorphic relations
title_full_unstemmed Enhancing supervised classifications with metamorphic relations
title_short Enhancing supervised classifications with metamorphic relations
title_sort enhancing supervised classifications with metamorphic relations
url https://eprints.nottingham.ac.uk/53764/
https://eprints.nottingham.ac.uk/53764/