Meaning space at SemEval-2018 Task 10: Combining explicitly encoded knowledge with information extracted from word embeddings

Research output: Chapter in Book / Report / Conference proceedingConference contributionAcademicpeer-review

Abstract

This paper presents the two systems submitted
by the meaning space team in Task 10 of the
SemEval competition 2018 entitled Capturing
discriminative attributes. The systems consist
of combinations of approaches exploiting explicitly
encoded knowledge about concepts in
WordNet and information encoded in distributional
semantic vectors. Rather than aiming
for high performance, we explore which
kind of semantic knowledge is best captured
by different methods. The results indicate that
WordNet glosses on different levels of the hierarchy
capture many attributes relevant for this
task. In combination with exploiting word embedding
similarities, this source of information
yielded our best results. Our best performing
system ranked 5th out of 13 final ranks. Our
analysis yields insights into the different kinds
of attributes represented by different sources
of knowledge.
LanguageEnglish
Title of host publicationProceedings of the 12th International Workshop on Semantic Evaluation (SemEval-2018)
Place of PublicationNew Orleans, United States
Pages940
Number of pages7
StatePublished - 2018

Fingerprint

Semantics

Cite this

Sommerauer, P. J. M., Fokkens, A. S., & Vossen, P. T. J. M. (2018). Meaning space at SemEval-2018 Task 10: Combining explicitly encoded knowledge with information extracted from word embeddings. In Proceedings of the 12th International Workshop on Semantic Evaluation (SemEval-2018) (pp. 940). New Orleans, United States.
Sommerauer, P.J.M. ; Fokkens, A.S. ; Vossen, P.T.J.M./ Meaning space at SemEval-2018 Task 10: Combining explicitly encoded knowledge with information extracted from word embeddings. Proceedings of the 12th International Workshop on Semantic Evaluation (SemEval-2018). New Orleans, United States, 2018. pp. 940
@inproceedings{8442e638c8604005b8a1c461b7215272,
title = "Meaning space at SemEval-2018 Task 10: Combining explicitly encoded knowledge with information extracted from word embeddings",
abstract = "This paper presents the two systems submittedby the meaning space team in Task 10 of theSemEval competition 2018 entitled Capturingdiscriminative attributes. The systems consistof combinations of approaches exploiting explicitlyencoded knowledge about concepts inWordNet and information encoded in distributionalsemantic vectors. Rather than aimingfor high performance, we explore whichkind of semantic knowledge is best capturedby different methods. The results indicate thatWordNet glosses on different levels of the hierarchycapture many attributes relevant for thistask. In combination with exploiting word embeddingsimilarities, this source of informationyielded our best results. Our best performingsystem ranked 5th out of 13 final ranks. Ouranalysis yields insights into the different kindsof attributes represented by different sourcesof knowledge.",
author = "P.J.M. Sommerauer and A.S. Fokkens and P.T.J.M. Vossen",
year = "2018",
language = "English",
pages = "940",
booktitle = "Proceedings of the 12th International Workshop on Semantic Evaluation (SemEval-2018)",

}

Sommerauer, PJM, Fokkens, AS & Vossen, PTJM 2018, Meaning space at SemEval-2018 Task 10: Combining explicitly encoded knowledge with information extracted from word embeddings. in Proceedings of the 12th International Workshop on Semantic Evaluation (SemEval-2018). New Orleans, United States, pp. 940.

Meaning space at SemEval-2018 Task 10: Combining explicitly encoded knowledge with information extracted from word embeddings. / Sommerauer, P.J.M.; Fokkens, A.S.; Vossen, P.T.J.M.

Proceedings of the 12th International Workshop on Semantic Evaluation (SemEval-2018). New Orleans, United States, 2018. p. 940.

Research output: Chapter in Book / Report / Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Meaning space at SemEval-2018 Task 10: Combining explicitly encoded knowledge with information extracted from word embeddings

AU - Sommerauer,P.J.M.

AU - Fokkens,A.S.

AU - Vossen,P.T.J.M.

PY - 2018

Y1 - 2018

N2 - This paper presents the two systems submittedby the meaning space team in Task 10 of theSemEval competition 2018 entitled Capturingdiscriminative attributes. The systems consistof combinations of approaches exploiting explicitlyencoded knowledge about concepts inWordNet and information encoded in distributionalsemantic vectors. Rather than aimingfor high performance, we explore whichkind of semantic knowledge is best capturedby different methods. The results indicate thatWordNet glosses on different levels of the hierarchycapture many attributes relevant for thistask. In combination with exploiting word embeddingsimilarities, this source of informationyielded our best results. Our best performingsystem ranked 5th out of 13 final ranks. Ouranalysis yields insights into the different kindsof attributes represented by different sourcesof knowledge.

AB - This paper presents the two systems submittedby the meaning space team in Task 10 of theSemEval competition 2018 entitled Capturingdiscriminative attributes. The systems consistof combinations of approaches exploiting explicitlyencoded knowledge about concepts inWordNet and information encoded in distributionalsemantic vectors. Rather than aimingfor high performance, we explore whichkind of semantic knowledge is best capturedby different methods. The results indicate thatWordNet glosses on different levels of the hierarchycapture many attributes relevant for thistask. In combination with exploiting word embeddingsimilarities, this source of informationyielded our best results. Our best performingsystem ranked 5th out of 13 final ranks. Ouranalysis yields insights into the different kindsof attributes represented by different sourcesof knowledge.

M3 - Conference contribution

SP - 940

BT - Proceedings of the 12th International Workshop on Semantic Evaluation (SemEval-2018)

CY - New Orleans, United States

ER -

Sommerauer PJM, Fokkens AS, Vossen PTJM. Meaning space at SemEval-2018 Task 10: Combining explicitly encoded knowledge with information extracted from word embeddings. In Proceedings of the 12th International Workshop on Semantic Evaluation (SemEval-2018). New Orleans, United States. 2018. p. 940.