Enhancing Supervised Classifications with Metamorphic Relations

Liming Xu; Dave Towey; Andrew P. French; Steve Benford; Zhi Quan Zhou; Tsong Yueh Chen

doi:10.1145/3193977.3193978

Enhancing Supervised Classifications with Metamorphic Relations

Liming Xu
, Dave Towey^*
, Andrew P. French
, Steve Benford
, Zhi Quan Zhou
, Tsong Yueh Chen

^*Corresponding author for this work

School of Computer Science

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

10 Citations (Scopus)

Abstract

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.

Original language	English
Title of host publication	Proceedings 2018 ACM/IEEE 3rd International Workshop on Metamorphic Testing, MET 2018
Publisher	IEEE Computer Society
Pages	46-53
Number of pages	8
ISBN (Electronic)	9781450357296
DOIs	https://doi.org/10.1145/3193977.3193978
Publication status	Published - 27 May 2018
Event	3rd ACM/IEEE International Workshop on Metamorphic Testing, MET 2018, held in conjunction with the 40th International Conference on Software Engineering, ICSE 2018 - Gothenburg, Sweden Duration: 27 May 2018 → …

Publication series

Name	Proceedings - International Conference on Software Engineering
ISSN (Print)	0270-5257

Conference

Conference	3rd ACM/IEEE International Workshop on Metamorphic Testing, MET 2018, held in conjunction with the 40th International Conference on Software Engineering, ICSE 2018
Country/Territory	Sweden
City	Gothenburg
Period	27/05/18 → …

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 15 Life on Land

Free Keywords

Artcodes
Metamorphic testing
metamorphic relations
random forests
supervised classification

ASJC Scopus subject areas

Software

Access to Document

10.1145/3193977.3193978

Cite this

Xu, L., Towey, D., French, A. P., Benford, S., Zhou, Z. Q., & Chen, T. Y. (2018). Enhancing Supervised Classifications with Metamorphic Relations. In Proceedings 2018 ACM/IEEE 3rd International Workshop on Metamorphic Testing, MET 2018 (pp. 46-53). (Proceedings - International Conference on Software Engineering). IEEE Computer Society. https://doi.org/10.1145/3193977.3193978

@inproceedings{d6b387dd52c842c18d0eb33e941fbdb1,

title = "Enhancing Supervised Classifications with Metamorphic Relations",

abstract = "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.",

keywords = "Artcodes, Metamorphic testing, metamorphic relations, random forests, supervised classification",

author = "Liming Xu and Dave Towey and French, \{Andrew P.\} and Steve Benford and Zhou, \{Zhi Quan\} and Chen, \{Tsong Yueh\}",

note = "Publisher Copyright: {\textcopyright} 2018 Association for Computing Machinery.; 3rd ACM/IEEE International Workshop on Metamorphic Testing, MET 2018, held in conjunction with the 40th International Conference on Software Engineering, ICSE 2018 ; Conference date: 27-05-2018",

year = "2018",

month = may,

day = "27",

doi = "10.1145/3193977.3193978",

language = "English",

series = "Proceedings - International Conference on Software Engineering",

publisher = "IEEE Computer Society",

pages = "46--53",

booktitle = "Proceedings 2018 ACM/IEEE 3rd International Workshop on Metamorphic Testing, MET 2018",

address = "United States",

}

Xu, L, Towey, D, French, AP, Benford, S, Zhou, ZQ & Chen, TY 2018, Enhancing Supervised Classifications with Metamorphic Relations. in Proceedings 2018 ACM/IEEE 3rd International Workshop on Metamorphic Testing, MET 2018. Proceedings - International Conference on Software Engineering, IEEE Computer Society, pp. 46-53, 3rd ACM/IEEE International Workshop on Metamorphic Testing, MET 2018, held in conjunction with the 40th International Conference on Software Engineering, ICSE 2018, Gothenburg, Sweden, 27/05/18. https://doi.org/10.1145/3193977.3193978

Enhancing Supervised Classifications with Metamorphic Relations. / Xu, Liming; Towey, Dave; French, Andrew P. et al.
Proceedings 2018 ACM/IEEE 3rd International Workshop on Metamorphic Testing, MET 2018. IEEE Computer Society, 2018. p. 46-53 (Proceedings - International Conference on Software Engineering).

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Enhancing Supervised Classifications with Metamorphic Relations

AU - Xu, Liming

AU - Towey, Dave

AU - French, Andrew P.

AU - Benford, Steve

AU - Zhou, Zhi Quan

AU - Chen, Tsong Yueh

PY - 2018/5/27

Y1 - 2018/5/27

N2 - 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.

AB - 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.

KW - Artcodes

KW - Metamorphic testing

KW - metamorphic relations

KW - random forests

KW - supervised classification

UR - https://www.scopus.com/pages/publications/85051271168

U2 - 10.1145/3193977.3193978

DO - 10.1145/3193977.3193978

M3 - Conference contribution

AN - SCOPUS:85051271168

T3 - Proceedings - International Conference on Software Engineering

SP - 46

EP - 53

BT - Proceedings 2018 ACM/IEEE 3rd International Workshop on Metamorphic Testing, MET 2018

PB - IEEE Computer Society

T2 - 3rd ACM/IEEE International Workshop on Metamorphic Testing, MET 2018, held in conjunction with the 40th International Conference on Software Engineering, ICSE 2018

Y2 - 27 May 2018

ER -

Enhancing Supervised Classifications with Metamorphic Relations

Abstract

Publication series

Conference

UN SDGs

Free Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this