Neural Models of Selectional Preferences for Implicit Semantic Role Labeling

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

Abstract

Implicit Semantic Role Labeling is a challenging task: it requires high-level understanding of the text while annotated data is very
limited. Due to the lack of training data, most researches either resort to simplistic machine learning methods or focus on automatically
acquiring training data. In this paper, we explore the possibilities of using more complex and expressive machine learning models
trained on a large amount of explicit roles. In addition, we compare the impact of one-way and multi-way selectional preference with the
hypothesis that the added information in multi-way models are beneficial. Although our models surpass a baseline that uses prototypical
vectors for SemEval-2010, we otherwise face mostly negative results. Selectional preference models perform lower than the baseline
on ON5V, a dataset of five ambiguous and frequent verbs. They are also outperformed by the Naive Bayes model of Feizabadi and
Pado (2015) on both datasets. We conclude that, even though multi-way selectional preference improves results for predicting explicit
semantic roles compared to one-way selectional preference, it harms performance for implicit roles. We release our source code,
including the reimplementation of two previously unavailable systems to enable further experimentation.
Original languageEnglish
Title of host publicationProceedings of the 11th edition of the Language Resources and Evaluation Conference (LREC2018)
EditorsHitoshi Isahara, Bente Maegaard, Stelios Piperidis, Christopher Cieri, Thierry Declerck, Koiti Hasida, Helene Mazo, Khalid Choukri, Sara Goggi, Joseph Mariani, Asuncion Moreno, Nicoletta Calzolari, Jan Odijk, Takenobu Tokunaga
Place of PublicationMiyazaki
PublisherEuropean Language Resources Association (ELRA)
Pages3062-3067
Number of pages6
ISBN (Electronic)9791095546009
ISBN (Print)9791095546009
Publication statusPublished - May 2018
EventInternational FrameNet Workshop at LREC, May 12, 2018 - Miyazaki, Japan
Duration: 12 May 201812 May 2018
Conference number: 3

Conference

ConferenceInternational FrameNet Workshop at LREC, May 12, 2018
Abbreviated titleLREC
CountryJapan
CityMiyazaki
Period12/05/1812/05/18

Fingerprint

semantics
learning method
Semantic Roles
Labeling
lack
learning
performance
Machine Learning
Expressive
Bayes Model
Verbs
Harm
Experimentation

Keywords

  • Implicit semantic role labeling
  • Neural network
  • Selectional preferences

Cite this

Le, M. N., & Fokkens, A. S. (2018). Neural Models of Selectional Preferences for Implicit Semantic Role Labeling. In H. Isahara, B. Maegaard, S. Piperidis, C. Cieri, T. Declerck, K. Hasida, H. Mazo, K. Choukri, S. Goggi, J. Mariani, A. Moreno, N. Calzolari, J. Odijk, ... T. Tokunaga (Eds.), Proceedings of the 11th edition of the Language Resources and Evaluation Conference (LREC2018) (pp. 3062-3067). [624] Miyazaki: European Language Resources Association (ELRA).
Le, M.N. ; Fokkens, A.S. / Neural Models of Selectional Preferences for Implicit Semantic Role Labeling. Proceedings of the 11th edition of the Language Resources and Evaluation Conference (LREC2018). editor / Hitoshi Isahara ; Bente Maegaard ; Stelios Piperidis ; Christopher Cieri ; Thierry Declerck ; Koiti Hasida ; Helene Mazo ; Khalid Choukri ; Sara Goggi ; Joseph Mariani ; Asuncion Moreno ; Nicoletta Calzolari ; Jan Odijk ; Takenobu Tokunaga. Miyazaki : European Language Resources Association (ELRA), 2018. pp. 3062-3067
@inproceedings{1affb53ab3314afe98160a7d2980c117,
title = "Neural Models of Selectional Preferences for Implicit Semantic Role Labeling",
abstract = "Implicit Semantic Role Labeling is a challenging task: it requires high-level understanding of the text while annotated data is verylimited. Due to the lack of training data, most researches either resort to simplistic machine learning methods or focus on automaticallyacquiring training data. In this paper, we explore the possibilities of using more complex and expressive machine learning modelstrained on a large amount of explicit roles. In addition, we compare the impact of one-way and multi-way selectional preference with thehypothesis that the added information in multi-way models are beneficial. Although our models surpass a baseline that uses prototypicalvectors for SemEval-2010, we otherwise face mostly negative results. Selectional preference models perform lower than the baselineon ON5V, a dataset of five ambiguous and frequent verbs. They are also outperformed by the Naive Bayes model of Feizabadi andPado (2015) on both datasets. We conclude that, even though multi-way selectional preference improves results for predicting explicitsemantic roles compared to one-way selectional preference, it harms performance for implicit roles. We release our source code,including the reimplementation of two previously unavailable systems to enable further experimentation.",
keywords = "Implicit semantic role labeling, Neural network, Selectional preferences",
author = "M.N. Le and A.S. Fokkens",
year = "2018",
month = "5",
language = "English",
isbn = "9791095546009",
pages = "3062--3067",
editor = "Hitoshi Isahara and Bente Maegaard and Stelios Piperidis and Christopher Cieri and Thierry Declerck and Koiti Hasida and Helene Mazo and Khalid Choukri and Sara Goggi and Joseph Mariani and Asuncion Moreno and Nicoletta Calzolari and Jan Odijk and Takenobu Tokunaga",
booktitle = "Proceedings of the 11th edition of the Language Resources and Evaluation Conference (LREC2018)",
publisher = "European Language Resources Association (ELRA)",

}

Le, MN & Fokkens, AS 2018, Neural Models of Selectional Preferences for Implicit Semantic Role Labeling. in H Isahara, B Maegaard, S Piperidis, C Cieri, T Declerck, K Hasida, H Mazo, K Choukri, S Goggi, J Mariani, A Moreno, N Calzolari, J Odijk & T Tokunaga (eds), Proceedings of the 11th edition of the Language Resources and Evaluation Conference (LREC2018)., 624, European Language Resources Association (ELRA), Miyazaki, pp. 3062-3067, International FrameNet Workshop at LREC, May 12, 2018, Miyazaki, Japan, 12/05/18.

Neural Models of Selectional Preferences for Implicit Semantic Role Labeling. / Le, M.N.; Fokkens, A.S.

Proceedings of the 11th edition of the Language Resources and Evaluation Conference (LREC2018). ed. / Hitoshi Isahara; Bente Maegaard; Stelios Piperidis; Christopher Cieri; Thierry Declerck; Koiti Hasida; Helene Mazo; Khalid Choukri; Sara Goggi; Joseph Mariani; Asuncion Moreno; Nicoletta Calzolari; Jan Odijk; Takenobu Tokunaga. Miyazaki : European Language Resources Association (ELRA), 2018. p. 3062-3067 624.

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

TY - GEN

T1 - Neural Models of Selectional Preferences for Implicit Semantic Role Labeling

AU - Le, M.N.

AU - Fokkens, A.S.

PY - 2018/5

Y1 - 2018/5

N2 - Implicit Semantic Role Labeling is a challenging task: it requires high-level understanding of the text while annotated data is verylimited. Due to the lack of training data, most researches either resort to simplistic machine learning methods or focus on automaticallyacquiring training data. In this paper, we explore the possibilities of using more complex and expressive machine learning modelstrained on a large amount of explicit roles. In addition, we compare the impact of one-way and multi-way selectional preference with thehypothesis that the added information in multi-way models are beneficial. Although our models surpass a baseline that uses prototypicalvectors for SemEval-2010, we otherwise face mostly negative results. Selectional preference models perform lower than the baselineon ON5V, a dataset of five ambiguous and frequent verbs. They are also outperformed by the Naive Bayes model of Feizabadi andPado (2015) on both datasets. We conclude that, even though multi-way selectional preference improves results for predicting explicitsemantic roles compared to one-way selectional preference, it harms performance for implicit roles. We release our source code,including the reimplementation of two previously unavailable systems to enable further experimentation.

AB - Implicit Semantic Role Labeling is a challenging task: it requires high-level understanding of the text while annotated data is verylimited. Due to the lack of training data, most researches either resort to simplistic machine learning methods or focus on automaticallyacquiring training data. In this paper, we explore the possibilities of using more complex and expressive machine learning modelstrained on a large amount of explicit roles. In addition, we compare the impact of one-way and multi-way selectional preference with thehypothesis that the added information in multi-way models are beneficial. Although our models surpass a baseline that uses prototypicalvectors for SemEval-2010, we otherwise face mostly negative results. Selectional preference models perform lower than the baselineon ON5V, a dataset of five ambiguous and frequent verbs. They are also outperformed by the Naive Bayes model of Feizabadi andPado (2015) on both datasets. We conclude that, even though multi-way selectional preference improves results for predicting explicitsemantic roles compared to one-way selectional preference, it harms performance for implicit roles. We release our source code,including the reimplementation of two previously unavailable systems to enable further experimentation.

KW - Implicit semantic role labeling

KW - Neural network

KW - Selectional preferences

UR - http://www.scopus.com/inward/record.url?scp=85059884548&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059884548&partnerID=8YFLogxK

UR - http://www.lrec-conf.org/proceedings/lrec2018/index.html

M3 - Conference contribution

SN - 9791095546009

SP - 3062

EP - 3067

BT - Proceedings of the 11th edition of the Language Resources and Evaluation Conference (LREC2018)

A2 - Isahara, Hitoshi

A2 - Maegaard, Bente

A2 - Piperidis, Stelios

A2 - Cieri, Christopher

A2 - Declerck, Thierry

A2 - Hasida, Koiti

A2 - Mazo, Helene

A2 - Choukri, Khalid

A2 - Goggi, Sara

A2 - Mariani, Joseph

A2 - Moreno, Asuncion

A2 - Calzolari, Nicoletta

A2 - Odijk, Jan

A2 - Tokunaga, Takenobu

PB - European Language Resources Association (ELRA)

CY - Miyazaki

ER -

Le MN, Fokkens AS. Neural Models of Selectional Preferences for Implicit Semantic Role Labeling. In Isahara H, Maegaard B, Piperidis S, Cieri C, Declerck T, Hasida K, Mazo H, Choukri K, Goggi S, Mariani J, Moreno A, Calzolari N, Odijk J, Tokunaga T, editors, Proceedings of the 11th edition of the Language Resources and Evaluation Conference (LREC2018). Miyazaki: European Language Resources Association (ELRA). 2018. p. 3062-3067. 624