Geometric Probing of Word Vectors

Madina Babazhanova, Maxat Tezekbayev, Zhenisbek Assylbekov

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

This paper studies the informativeness of linguistic properties such as part-of-speech and named entities encoded in word representations. First, we find directions that correspond to these properties using the method of Elazar et al. (2020). Then such directions are compared with the principal vectors obtained from application of PCA to word embeddings. As a result, we find that the part-of-speech information is more important for word embeddings than the named entity property.

Original languageEnglish
Title of host publicationESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
Publisheri6doc.com publication
Pages587-592
Number of pages6
ISBN (Electronic)9782875870827
DOIs
Publication statusPublished - 2021
Event29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2021 - Virtual, Online, Belgium
Duration: Oct 6 2021Oct 8 2021

Publication series

NameESANN 2021 Proceedings - 29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

Conference

Conference29th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2021
Country/TerritoryBelgium
CityVirtual, Online
Period10/6/2110/8/21

ASJC Scopus subject areas

  • Artificial Intelligence
  • Information Systems

Fingerprint

Dive into the research topics of 'Geometric Probing of Word Vectors'. Together they form a unique fingerprint.

Cite this