Graded hyponymy for compositional distributional semantics

The categorical compositional distributional model of natural language provides a conceptually motivated procedure to compute the meaning of a sentence, given its grammatical structure and the meanings of its words. This approach has outperformed other models in mainstream empirical language process...

Full description

Bibliographic Details
Main Authors: Dea Bankova, Bob Coecke, Martha Lewis, Dan Marsden
Format: Article
Language:English
Published: Institute of Computer Science, Polish Academy of Sciences 2019-03-01
Series:Journal of Language Modelling
Subjects:
Online Access:https://jlm.ipipan.waw.pl/index.php/JLM/article/view/230
_version_ 1798030582587850752
author Dea Bankova
Bob Coecke
Martha Lewis
Dan Marsden
author_facet Dea Bankova
Bob Coecke
Martha Lewis
Dan Marsden
author_sort Dea Bankova
collection DOAJ
description The categorical compositional distributional model of natural language provides a conceptually motivated procedure to compute the meaning of a sentence, given its grammatical structure and the meanings of its words. This approach has outperformed other models in mainstream empirical language processing tasks, but lacks an effective model of lexical entailment. We address this shortcoming by exploiting the freedom in our abstract categorical framework to change our choice of semantic model. This allows us to describe hyponymy as a graded order on meanings, using models of partial information used in quantum computation. Quantum logic embeds in this graded order.
first_indexed 2024-04-11T19:42:38Z
format Article
id doaj.art-48d5afed5c254d47b58f9dd7e3806bec
institution Directory Open Access Journal
issn 2299-856X
2299-8470
language English
last_indexed 2024-04-11T19:42:38Z
publishDate 2019-03-01
publisher Institute of Computer Science, Polish Academy of Sciences
record_format Article
series Journal of Language Modelling
spelling doaj.art-48d5afed5c254d47b58f9dd7e3806bec2022-12-22T04:06:37ZengInstitute of Computer Science, Polish Academy of SciencesJournal of Language Modelling2299-856X2299-84702019-03-0162225–260225–26010.15398/jlm.v6i2.230171Graded hyponymy for compositional distributional semanticsDea Bankova0Bob Coecke1Martha Lewis2Dan Marsden3University of Oxford, DataSineDepartment of Computer Science, University of OxfordUniversity of AmsterdamDepartment of Computer Science, University of OxfordThe categorical compositional distributional model of natural language provides a conceptually motivated procedure to compute the meaning of a sentence, given its grammatical structure and the meanings of its words. This approach has outperformed other models in mainstream empirical language processing tasks, but lacks an effective model of lexical entailment. We address this shortcoming by exploiting the freedom in our abstract categorical framework to change our choice of semantic model. This allows us to describe hyponymy as a graded order on meanings, using models of partial information used in quantum computation. Quantum logic embeds in this graded order.https://jlm.ipipan.waw.pl/index.php/JLM/article/view/230distributional semanticshyponymycategorical composition
spellingShingle Dea Bankova
Bob Coecke
Martha Lewis
Dan Marsden
Graded hyponymy for compositional distributional semantics
Journal of Language Modelling
distributional semantics
hyponymy
categorical composition
title Graded hyponymy for compositional distributional semantics
title_full Graded hyponymy for compositional distributional semantics
title_fullStr Graded hyponymy for compositional distributional semantics
title_full_unstemmed Graded hyponymy for compositional distributional semantics
title_short Graded hyponymy for compositional distributional semantics
title_sort graded hyponymy for compositional distributional semantics
topic distributional semantics
hyponymy
categorical composition
url https://jlm.ipipan.waw.pl/index.php/JLM/article/view/230
work_keys_str_mv AT deabankova gradedhyponymyforcompositionaldistributionalsemantics
AT bobcoecke gradedhyponymyforcompositionaldistributionalsemantics
AT marthalewis gradedhyponymyforcompositionaldistributionalsemantics
AT danmarsden gradedhyponymyforcompositionaldistributionalsemantics