Automatic classification of containment and support spatial relations in English and Dutch

Kate Lockwood*, Andrew Lovett, Ken Forbus

*Corresponding author for this work

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

7 Scopus citations

Abstract

The need to communicate and reason about space is pervasive in human cognition. Consequently, most languages develop specialized terms for describing relationships between objects in space - spatial prepositions. However, the specific set of prepositions and the delineations between them vary widely. For example, in English containment relationships are categorized as in and support relationships are classified as on. In Dutch, on the other hand, three different prepositions are used to distinguish between different types of support relations: op, aan, and om. In this paper we show how progressive alignment can be used to model the formation of spatial language categories along the containment-support continuum in both English and Dutch.

Original languageEnglish (US)
Title of host publicationSpatial Cognition VI
Subtitle of host publicationLearning, Reasoning, and Talking about Space - International Conference Spatial Cognition 2008, Proceedings
Pages283-294
Number of pages12
DOIs
StatePublished - 2008
EventInternational Conference Spatial Cognition 2008 - Freiburg, Germany
Duration: Sep 15 2008Sep 19 2008

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5248 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherInternational Conference Spatial Cognition 2008
Country/TerritoryGermany
CityFreiburg
Period9/15/089/19/08

Keywords

  • Cognitive modeling
  • Spatial prepositions

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Automatic classification of containment and support spatial relations in English and Dutch'. Together they form a unique fingerprint.

Cite this