Bayesian Word Alignment for Massively Parallel Texts

Östling, Robert

doi:10.3115/v1/e14-4024

Search citation statements

Order By: Relevance

Paper Sections

Select...

Entropy In Distributional Semantics1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2015

2021

Publication Types

Select...

Other6

Relationship

Self Cite1

Independent5

Authors

Journals

Cited by 8 publications

(1 citation statement)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These estimations are also relevant for efforts to broaden the scope of NLP to lesser known languages [14,21,23,7], and to establish quantitative and corpus-based methods in linguistic typology [6,32,31].…”

Section: Entropy In Distributional Semanticsmentioning

confidence: 99%

The word entropy of natural languages

Bentz,

Alikaniotis

2016

Preprint

View full text Add to dashboard Cite

The average uncertainty associated with words is an informationtheoretic concept at the heart of quantitative and computational linguistics. The entropy has been established as a measure of this average uncertainty -also called average information content. We here use parallel texts of 21 languages to establish the number of tokens at which word entropies converge to stable values. These convergence points are then used to select texts from a massively parallel corpus, and to estimate word entropies across more than 1000 languages. Our results help to establish quantitative language comparisons, to understand the performance of multilingual translation systems, and to normalize semantic similarity measures.

show abstract