A study of textual entailment

Vasile Rus, Philip M. McCarthy, Danielle McNamara, Arthur C. Graesser

Research output: Contribution to journalArticle

23 Citations (Scopus)

Abstract

In this paper we study a graph-based approach to the task of Recognizing Textual Entailment between a Text and a Hypothesis. The approach takes into account the full lexico-syntactic context of both the Text and Hypothesis and is based on the concept of subsumption. It starts with mapping the Text and Hypothesis on to graph structures that have nodes representing concepts and edges representing lexico-syntactic relations among concepts. An entailment decision is then made on the basis of a subsumption score between the Text-graph and Hypothesis-graph. The results obtained from a standard entailment test data set were promising. The impact of synonymy on entailment is quantified and discussed. An important advantage to a solution like ours is its ability to be customized to obtain high-confidence results.

Original languageEnglish (US)
Pages (from-to)659-685
Number of pages27
JournalInternational Journal on Artificial Intelligence Tools
Volume17
Issue number4
DOIs
StatePublished - Aug 2008
Externally publishedYes

Fingerprint

Syntactics

Keywords

  • Graph subsumption
  • Natural language processing
  • Syntactic dependencies
  • Textual entailment

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

A study of textual entailment. / Rus, Vasile; McCarthy, Philip M.; McNamara, Danielle; Graesser, Arthur C.

In: International Journal on Artificial Intelligence Tools, Vol. 17, No. 4, 08.2008, p. 659-685.

Research output: Contribution to journalArticle

Rus, Vasile ; McCarthy, Philip M. ; McNamara, Danielle ; Graesser, Arthur C. / A study of textual entailment. In: International Journal on Artificial Intelligence Tools. 2008 ; Vol. 17, No. 4. pp. 659-685.
@article{0db5ea50e80d4c79b88f70a9f98701c8,
title = "A study of textual entailment",
abstract = "In this paper we study a graph-based approach to the task of Recognizing Textual Entailment between a Text and a Hypothesis. The approach takes into account the full lexico-syntactic context of both the Text and Hypothesis and is based on the concept of subsumption. It starts with mapping the Text and Hypothesis on to graph structures that have nodes representing concepts and edges representing lexico-syntactic relations among concepts. An entailment decision is then made on the basis of a subsumption score between the Text-graph and Hypothesis-graph. The results obtained from a standard entailment test data set were promising. The impact of synonymy on entailment is quantified and discussed. An important advantage to a solution like ours is its ability to be customized to obtain high-confidence results.",
keywords = "Graph subsumption, Natural language processing, Syntactic dependencies, Textual entailment",
author = "Vasile Rus and McCarthy, {Philip M.} and Danielle McNamara and Graesser, {Arthur C.}",
year = "2008",
month = "8",
doi = "10.1142/S0218213008004096",
language = "English (US)",
volume = "17",
pages = "659--685",
journal = "International Journal on Artificial Intelligence Tools",
issn = "0218-2130",
publisher = "World Scientific Publishing Co. Pte Ltd",
number = "4",

}

TY - JOUR

T1 - A study of textual entailment

AU - Rus, Vasile

AU - McCarthy, Philip M.

AU - McNamara, Danielle

AU - Graesser, Arthur C.

PY - 2008/8

Y1 - 2008/8

N2 - In this paper we study a graph-based approach to the task of Recognizing Textual Entailment between a Text and a Hypothesis. The approach takes into account the full lexico-syntactic context of both the Text and Hypothesis and is based on the concept of subsumption. It starts with mapping the Text and Hypothesis on to graph structures that have nodes representing concepts and edges representing lexico-syntactic relations among concepts. An entailment decision is then made on the basis of a subsumption score between the Text-graph and Hypothesis-graph. The results obtained from a standard entailment test data set were promising. The impact of synonymy on entailment is quantified and discussed. An important advantage to a solution like ours is its ability to be customized to obtain high-confidence results.

AB - In this paper we study a graph-based approach to the task of Recognizing Textual Entailment between a Text and a Hypothesis. The approach takes into account the full lexico-syntactic context of both the Text and Hypothesis and is based on the concept of subsumption. It starts with mapping the Text and Hypothesis on to graph structures that have nodes representing concepts and edges representing lexico-syntactic relations among concepts. An entailment decision is then made on the basis of a subsumption score between the Text-graph and Hypothesis-graph. The results obtained from a standard entailment test data set were promising. The impact of synonymy on entailment is quantified and discussed. An important advantage to a solution like ours is its ability to be customized to obtain high-confidence results.

KW - Graph subsumption

KW - Natural language processing

KW - Syntactic dependencies

KW - Textual entailment

UR - http://www.scopus.com/inward/record.url?scp=51549098376&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51549098376&partnerID=8YFLogxK

U2 - 10.1142/S0218213008004096

DO - 10.1142/S0218213008004096

M3 - Article

VL - 17

SP - 659

EP - 685

JO - International Journal on Artificial Intelligence Tools

JF - International Journal on Artificial Intelligence Tools

SN - 0218-2130

IS - 4

ER -