Enhancing supervised classifications with metamorphic relations

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

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.

Citation

Xu, L., Towey, D., French, A. P., Benford, S., Zhou, Z. Q., & Chen, T. Y. (2018). Enhancing supervised classifications with metamorphic relations. In MET '18: Proceedings of the 3rd International Workshop on Metamorphic Testing (46-53). https://doi.org/10.1145/3193977.3193978

• Conference Name: Proceedings of the 3rd International Workshop on Metamorphic Testing - MET '18
• Start Date: May 27, 2018
• End Date: May 27, 2018
• Acceptance Date: Mar 4, 2018
• Publication Date: May 27, 2018
• Deposit Date: Aug 9, 2018
• Publicly Available Date: Aug 9, 2018
• Publisher: Association for Computing Machinery (ACM)
• Pages: 46-53
• Book Title: MET '18: Proceedings of the 3rd International Workshop on Metamorphic Testing
• ISBN: 9781450357296
• DOI: https://doi.org/10.1145/3193977.3193978
• Public URL: https://nottingham-repository.worktribe.com/output/986144
• Publisher URL: https://dl.acm.org/citation.cfm?doid=3193977.3193978