ReferenceNet: a semantic-pragmatic network for capturing reference relations

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

Abstract

In this paper, we present ReferenceNet: a semantic-pragmatic network of reference relations between synsets. Synonyms are assumed to be exchangeable in similar contexts and also word embeddings are based on sharing of local contexts represented as vectors. Co-referring words, however, tend to occur in the same topical context but in different local contexts. In addition, they may express different concepts related through topical coherence, and through author framing and perspective. In this paper, we describe how reference relations can be added to WordNet and how they can be acquired. We evaluate two methods of extracting event coreference relations using WordNet relations against a manual annotation of 38 documents within the same topical domain of gun violence. We conclude that precision is reasonable but recall is lower because the WordNet hierarchy does not sufficiently capture the required coherence and perspective relations.
LanguageUndefined/Unknown
Title of host publicationGlobal Wordnet Conference 2018, Singapore
StatePublished - 2018

Cite this

@inproceedings{13832531c83345a781ed0a51415d576b,
title = "ReferenceNet: a semantic-pragmatic network for capturing reference relations",
abstract = "In this paper, we present ReferenceNet: a semantic-pragmatic network of reference relations between synsets. Synonyms are assumed to be exchangeable in similar contexts and also word embeddings are based on sharing of local contexts represented as vectors. Co-referring words, however, tend to occur in the same topical context but in different local contexts. In addition, they may express different concepts related through topical coherence, and through author framing and perspective. In this paper, we describe how reference relations can be added to WordNet and how they can be acquired. We evaluate two methods of extracting event coreference relations using WordNet relations against a manual annotation of 38 documents within the same topical domain of gun violence. We conclude that precision is reasonable but recall is lower because the WordNet hierarchy does not sufficiently capture the required coherence and perspective relations.",
author = "Piek Vossen and Marten Postma and Filip Ilievski",
year = "2018",
language = "Undefined/Unknown",
booktitle = "Global Wordnet Conference 2018, Singapore",

}

ReferenceNet: a semantic-pragmatic network for capturing reference relations. / Vossen, Piek; Postma, Marten; Ilievski, Filip.

Global Wordnet Conference 2018, Singapore. 2018.

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

TY - GEN

T1 - ReferenceNet: a semantic-pragmatic network for capturing reference relations

AU - Vossen,Piek

AU - Postma,Marten

AU - Ilievski,Filip

PY - 2018

Y1 - 2018

N2 - In this paper, we present ReferenceNet: a semantic-pragmatic network of reference relations between synsets. Synonyms are assumed to be exchangeable in similar contexts and also word embeddings are based on sharing of local contexts represented as vectors. Co-referring words, however, tend to occur in the same topical context but in different local contexts. In addition, they may express different concepts related through topical coherence, and through author framing and perspective. In this paper, we describe how reference relations can be added to WordNet and how they can be acquired. We evaluate two methods of extracting event coreference relations using WordNet relations against a manual annotation of 38 documents within the same topical domain of gun violence. We conclude that precision is reasonable but recall is lower because the WordNet hierarchy does not sufficiently capture the required coherence and perspective relations.

AB - In this paper, we present ReferenceNet: a semantic-pragmatic network of reference relations between synsets. Synonyms are assumed to be exchangeable in similar contexts and also word embeddings are based on sharing of local contexts represented as vectors. Co-referring words, however, tend to occur in the same topical context but in different local contexts. In addition, they may express different concepts related through topical coherence, and through author framing and perspective. In this paper, we describe how reference relations can be added to WordNet and how they can be acquired. We evaluate two methods of extracting event coreference relations using WordNet relations against a manual annotation of 38 documents within the same topical domain of gun violence. We conclude that precision is reasonable but recall is lower because the WordNet hierarchy does not sufficiently capture the required coherence and perspective relations.

M3 - Conference contribution

BT - Global Wordnet Conference 2018, Singapore

ER -