Finding the best model among representative compositional models

Masayasu Muraoka, Sonse Shimaoka, Kazeto Yamamoto, Yotaro Watanabe, Naoaki Okazaki, Kentaro Inui

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

12 Citations (Scopus)

Abstract

The field of distributional-compositional semantics has yielded a range of computational models for composing the vector of a phrase from those of constituent word vectors. Existing models have various ranges of their expressiveness, recursivity, and trainability. However, these models have not been examined closely for their compositionality. We implement and compare these models under the same conditions. The experimentally obtained results demonstrate that the model using different composition matrices for different dependency relations achieved state-ofthe-art performance on a dataset for two-word compositions (Mitchell and Lapata, 2010).

Original languageEnglish
Title of host publicationProceedings of the 28th Pacific Asia Conference on Language, Information and Computation, PACLIC 2014
EditorsPrachya Boonkwan, Wirote Aroonmanakun, Thepchai Supnithi
PublisherFaculty of Pharmaceutical Sciences, Chulalongkorn University
Pages65-74
Number of pages10
ISBN (Electronic)9786165518871
Publication statusPublished - 2014
Event28th Pacific Asia Conference on Language, Information and Computation, PACLIC 2014 - Phuket, Thailand
Duration: 2014 Dec 122014 Dec 14

Publication series

NameProceedings of the 28th Pacific Asia Conference on Language, Information and Computation, PACLIC 2014

Other

Other28th Pacific Asia Conference on Language, Information and Computation, PACLIC 2014
Country/TerritoryThailand
CityPhuket
Period14/12/1214/12/14

ASJC Scopus subject areas

  • Language and Linguistics
  • Computer Science (miscellaneous)

Fingerprint

Dive into the research topics of 'Finding the best model among representative compositional models'. Together they form a unique fingerprint.

Cite this