Incorporating frame information to semantic role labeling

Joo Young Lee, Young In Song, Hae-Chang Rim, Kyoung Soo Han

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

In this paper, we suggest a new probabilistic model of semantic role labeling, which uses the frameset of the predicate as explicit linguistic knowledge for providing global information on the predicateargument structure that local classifier is unable to catch. The proposed model consists of three sub-models: role sequence generation model, frameset generation model, and matching model. The role sequence generation model generates the semantic role sequence candidates of a given predicate by using the local classification approach, which is a widely used approach in previous research. The frameset generation model estimates the probability of each frameset that the predicate can take. The matching model is designed to measure the degree of the matching between the generated role sequence and the frameset by using several features. These features are developed to represent the predicate-argument structure information described in the frameset. In the experiments our model shows that the use of knowledge about the predicate-argument structure is effective for selecting a more appropriate semantic role sequence.

Original languageEnglish
Pages (from-to)201-204
Number of pages4
JournalIEICE Transactions on Information and Systems
VolumeE93-D
Issue number1
Publication statusPublished - 2010 Apr 6

Fingerprint

Labeling
Semantics
Linguistics
Classifiers

Keywords

  • Frame information
  • Frameset
  • Predicate-argument structure
  • Propbank
  • Semantic role labeling

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Software
  • Artificial Intelligence
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition

Cite this

Incorporating frame information to semantic role labeling. / Lee, Joo Young; Song, Young In; Rim, Hae-Chang; Han, Kyoung Soo.

In: IEICE Transactions on Information and Systems, Vol. E93-D, No. 1, 06.04.2010, p. 201-204.

Research output: Contribution to journalArticle

Lee, Joo Young ; Song, Young In ; Rim, Hae-Chang ; Han, Kyoung Soo. / Incorporating frame information to semantic role labeling. In: IEICE Transactions on Information and Systems. 2010 ; Vol. E93-D, No. 1. pp. 201-204.
@article{d1b58833ad1b42b59b920346823b20cd,
title = "Incorporating frame information to semantic role labeling",
abstract = "In this paper, we suggest a new probabilistic model of semantic role labeling, which uses the frameset of the predicate as explicit linguistic knowledge for providing global information on the predicateargument structure that local classifier is unable to catch. The proposed model consists of three sub-models: role sequence generation model, frameset generation model, and matching model. The role sequence generation model generates the semantic role sequence candidates of a given predicate by using the local classification approach, which is a widely used approach in previous research. The frameset generation model estimates the probability of each frameset that the predicate can take. The matching model is designed to measure the degree of the matching between the generated role sequence and the frameset by using several features. These features are developed to represent the predicate-argument structure information described in the frameset. In the experiments our model shows that the use of knowledge about the predicate-argument structure is effective for selecting a more appropriate semantic role sequence.",
keywords = "Frame information, Frameset, Predicate-argument structure, Propbank, Semantic role labeling",
author = "Lee, {Joo Young} and Song, {Young In} and Hae-Chang Rim and Han, {Kyoung Soo}",
year = "2010",
month = "4",
day = "6",
language = "English",
volume = "E93-D",
pages = "201--204",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "1",

}

TY - JOUR

T1 - Incorporating frame information to semantic role labeling

AU - Lee, Joo Young

AU - Song, Young In

AU - Rim, Hae-Chang

AU - Han, Kyoung Soo

PY - 2010/4/6

Y1 - 2010/4/6

N2 - In this paper, we suggest a new probabilistic model of semantic role labeling, which uses the frameset of the predicate as explicit linguistic knowledge for providing global information on the predicateargument structure that local classifier is unable to catch. The proposed model consists of three sub-models: role sequence generation model, frameset generation model, and matching model. The role sequence generation model generates the semantic role sequence candidates of a given predicate by using the local classification approach, which is a widely used approach in previous research. The frameset generation model estimates the probability of each frameset that the predicate can take. The matching model is designed to measure the degree of the matching between the generated role sequence and the frameset by using several features. These features are developed to represent the predicate-argument structure information described in the frameset. In the experiments our model shows that the use of knowledge about the predicate-argument structure is effective for selecting a more appropriate semantic role sequence.

AB - In this paper, we suggest a new probabilistic model of semantic role labeling, which uses the frameset of the predicate as explicit linguistic knowledge for providing global information on the predicateargument structure that local classifier is unable to catch. The proposed model consists of three sub-models: role sequence generation model, frameset generation model, and matching model. The role sequence generation model generates the semantic role sequence candidates of a given predicate by using the local classification approach, which is a widely used approach in previous research. The frameset generation model estimates the probability of each frameset that the predicate can take. The matching model is designed to measure the degree of the matching between the generated role sequence and the frameset by using several features. These features are developed to represent the predicate-argument structure information described in the frameset. In the experiments our model shows that the use of knowledge about the predicate-argument structure is effective for selecting a more appropriate semantic role sequence.

KW - Frame information

KW - Frameset

KW - Predicate-argument structure

KW - Propbank

KW - Semantic role labeling

UR - http://www.scopus.com/inward/record.url?scp=77950228703&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77950228703&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:77950228703

VL - E93-D

SP - 201

EP - 204

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 1

ER -