Data augmentation for hypernymy detection

Kober, Thomas, Weeds, Julie, Bertolini, Lorenzo and David, Weir (2021) Data augmentation for hypernymy detection. 16th conference of the European Chapter of the Association for Computational Linguistics (EACL), Kyiv / online, 19 - 23 April, 2021. Published in: Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume. 1034-1048. Association for Computational Linguistics

[img] PDF - Accepted Version
Available under License Creative Commons Attribution.

Download (1MB)


The automatic detection of hypernymy relationships represents a challenging problem in NLP. The successful application of state-of-the-art supervised approaches using distributed representations has generally been impeded by the limited availability of high quality training data. We have developed two novel data augmentation techniques which generate new training examples from existing ones. First, we combine the linguistic principles of hypernym transitivity and intersective modifier-noun composition to generate additional pairs of vectors, such as “small dog - dog” or “small dog - animal”, for which a hypernymy relationship can be assumed. Second, we use generative adversarial networks (GANs) to generate pairs of vectors for which the hypernymy relation can also be assumed. We furthermore present two complementary strategies for extending an existing dataset by leveraging linguistic resources such as WordNet. Using an evaluation across 3 different datasets for hypernymy detection and 2 different vector spaces, we demonstrate that both of the proposed automatic data augmentation and dataset extension strategies substantially improve classifier performance.

Item Type: Conference Proceedings
Schools and Departments: School of Engineering and Informatics > Informatics
Related URLs:
Depositing User: David Weir
Date Deposited: 03 Feb 2021 10:46
Last Modified: 28 Apr 2021 07:06

View download statistics for this item

📧 Request an update