Edinburgh Research Explorer

Improving Semantic Composition with Offset Inference

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

Related Edinburgh Organisations

Open Access permissions

Open

Original languageEnglish
Title of host publicationProceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
PublisherAssociation for Computational Linguistics
Pages433-440
Number of pages8
DOIs
Publication statusPublished - 30 Jul 2017
Event55th annual meeting of the Association for Computational Linguistics (ACL) - Vancouver, Canada
Duration: 30 Jul 20174 Aug 2017
http://acl2017.org/

Conference

Conference55th annual meeting of the Association for Computational Linguistics (ACL)
Abbreviated titleACL 2017
CountryCanada
CityVancouver
Period30/07/174/08/17
Internet address

Abstract

Count-based distributional semantic models suffer from sparsity due to unobserved but plausible co-occurrences in any text collection. This problem is amplified for models like Anchored Packed Trees (APTs), that take the grammatical type of a co-occurrence into account. We therefore introduce a novel form of distributional inference that exploits the rich type structure in APTs and infers missing data by the same mechanism that is used for semantic composition.

Event

55th annual meeting of the Association for Computational Linguistics (ACL)

30/07/174/08/17

Vancouver, Canada

Event: Conference

Download statistics

No data available

ID: 55588368