Generating natural language descriptions using speaker-dependent information

Thiago Castro Ferreira, Ivandré Paraboni

Research output: Contribution to journalArticleScientificpeer-review

Abstract

This paper discusses the issue of human variation in natural language referring expression generation. We introduce a model of content selection that takes speaker-dependent information into account to produce descriptions that closely resemble those produced by each individual, as seen in a number of reference corpora. Results show that our speaker-dependent referring expression generation model outperforms alternatives that do not take human variation into account, or which do so less extensively, and suggest that the use of machine-learning methods may be an ideal approach to mimic complex referential behaviour.
Original languageEnglish
JournalNatural Language Engineering
DOIs
Publication statusPublished - 27 Feb 2017

Fingerprint

learning method
language
Learning systems
Natural Language
Referring Expressions
Referential
Machine Learning

Cite this

@article{69251893b51a431093203484b2392d32,
title = "Generating natural language descriptions using speaker-dependent information",
abstract = "This paper discusses the issue of human variation in natural language referring expression generation. We introduce a model of content selection that takes speaker-dependent information into account to produce descriptions that closely resemble those produced by each individual, as seen in a number of reference corpora. Results show that our speaker-dependent referring expression generation model outperforms alternatives that do not take human variation into account, or which do so less extensively, and suggest that the use of machine-learning methods may be an ideal approach to mimic complex referential behaviour.",
author = "{Castro Ferreira}, Thiago and Ivandr{\'e} Paraboni",
year = "2017",
month = "2",
day = "27",
doi = "10.1017/S1351324917000079",
language = "English",
journal = "Natural Language Engineering",
issn = "1351-3249",
publisher = "Cambridge University Press",

}

Generating natural language descriptions using speaker-dependent information. / Castro Ferreira, Thiago; Paraboni, Ivandré.

In: Natural Language Engineering, 27.02.2017.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Generating natural language descriptions using speaker-dependent information

AU - Castro Ferreira, Thiago

AU - Paraboni, Ivandré

PY - 2017/2/27

Y1 - 2017/2/27

N2 - This paper discusses the issue of human variation in natural language referring expression generation. We introduce a model of content selection that takes speaker-dependent information into account to produce descriptions that closely resemble those produced by each individual, as seen in a number of reference corpora. Results show that our speaker-dependent referring expression generation model outperforms alternatives that do not take human variation into account, or which do so less extensively, and suggest that the use of machine-learning methods may be an ideal approach to mimic complex referential behaviour.

AB - This paper discusses the issue of human variation in natural language referring expression generation. We introduce a model of content selection that takes speaker-dependent information into account to produce descriptions that closely resemble those produced by each individual, as seen in a number of reference corpora. Results show that our speaker-dependent referring expression generation model outperforms alternatives that do not take human variation into account, or which do so less extensively, and suggest that the use of machine-learning methods may be an ideal approach to mimic complex referential behaviour.

U2 - 10.1017/S1351324917000079

DO - 10.1017/S1351324917000079

M3 - Article

JO - Natural Language Engineering

JF - Natural Language Engineering

SN - 1351-3249

ER -