Dialogue-act tagging using smart feature selection: results on multiple corpora

Daan Verbree; R.J. Rienks; Dirk K.J. Heylen

doi:10.1109/SLT.2006.326819

Dialogue-act tagging using smart feature selection: results on multiple corpora

Daan Verbree, R.J. Rienks, Dirk K.J. Heylen

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

27 Citations (Scopus)

240 Downloads (Pure)

Abstract

This paper presents an overview of our on-going work on dialogueact classification. Results are presented on the ICSI, Switchboard, and on a selection of the AMI corpus, setting a baseline for forthcoming research. For these corpora the best accuracy scores obtained are 89.27%, 65.68% and 59.76%, respectively. We introduce a smart compression technique for feature selection and compare the performance from a subset of the AMI transcriptions with AMI-ASR output for the same subset.

Original language	Undefined
Title of host publication	First International IEEE Workshop on Spoken Language Technology SLT 2006
Editors	B. Raorke
Place of Publication	Palm Beach
Publisher	IEEE Computer Society
Pages	70-73
Number of pages	4
ISBN (Print)	1-4244-0873-3
DOIs	https://doi.org/10.1109/SLT.2006.326819
Publication status	Published - Dec 2006

Publication series

Name
Publisher	IEEE Computer Society
Number	10

Keywords

EC Grant Agreement nr.: FP6/506811
EWI-9501
IR-67003
METIS-237662
HMI-CI: Computational Intelligence

Access to Document

10.1109/SLT.2006.326819

Verbee
open
Verbee
open

Cite this

Verbree, D., Rienks, R. J., & Heylen, D. K. J. (2006). Dialogue-act tagging using smart feature selection: results on multiple corpora. In B. Raorke (Ed.), First International IEEE Workshop on Spoken Language Technology SLT 2006 (pp. 70-73). Palm Beach: IEEE Computer Society. https://doi.org/10.1109/SLT.2006.326819

@inproceedings{c2a7d4f29d144c70b1b9c31b7b43b9b3,

title = "Dialogue-act tagging using smart feature selection: results on multiple corpora",

abstract = "This paper presents an overview of our on-going work on dialogueact classification. Results are presented on the ICSI, Switchboard, and on a selection of the AMI corpus, setting a baseline for forthcoming research. For these corpora the best accuracy scores obtained are 89.27{\%}, 65.68{\%} and 59.76{\%}, respectively. We introduce a smart compression technique for feature selection and compare the performance from a subset of the AMI transcriptions with AMI-ASR output for the same subset.",

keywords = "EC Grant Agreement nr.: FP6/506811, EWI-9501, IR-67003, METIS-237662, HMI-CI: Computational Intelligence",

author = "Daan Verbree and R.J. Rienks and Heylen, {Dirk K.J.}",

year = "2006",

month = "12",

doi = "10.1109/SLT.2006.326819",

language = "Undefined",

isbn = "1-4244-0873-3",

publisher = "IEEE Computer Society",

number = "10",

pages = "70--73",

editor = "B. Raorke",

booktitle = "First International IEEE Workshop on Spoken Language Technology SLT 2006",

address = "United States",

}

Dialogue-act tagging using smart feature selection: results on multiple corpora. / Verbree, Daan; Rienks, R.J.; Heylen, Dirk K.J.

First International IEEE Workshop on Spoken Language Technology SLT 2006. ed. / B. Raorke. Palm Beach : IEEE Computer Society, 2006. p. 70-73.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Dialogue-act tagging using smart feature selection: results on multiple corpora

AU - Verbree, Daan

AU - Rienks, R.J.

AU - Heylen, Dirk K.J.

PY - 2006/12

Y1 - 2006/12

N2 - This paper presents an overview of our on-going work on dialogueact classification. Results are presented on the ICSI, Switchboard, and on a selection of the AMI corpus, setting a baseline for forthcoming research. For these corpora the best accuracy scores obtained are 89.27%, 65.68% and 59.76%, respectively. We introduce a smart compression technique for feature selection and compare the performance from a subset of the AMI transcriptions with AMI-ASR output for the same subset.

AB - This paper presents an overview of our on-going work on dialogueact classification. Results are presented on the ICSI, Switchboard, and on a selection of the AMI corpus, setting a baseline for forthcoming research. For these corpora the best accuracy scores obtained are 89.27%, 65.68% and 59.76%, respectively. We introduce a smart compression technique for feature selection and compare the performance from a subset of the AMI transcriptions with AMI-ASR output for the same subset.

KW - EC Grant Agreement nr.: FP6/506811

KW - EWI-9501

KW - IR-67003

KW - METIS-237662

KW - HMI-CI: Computational Intelligence

U2 - 10.1109/SLT.2006.326819

DO - 10.1109/SLT.2006.326819

M3 - Conference contribution

SN - 1-4244-0873-3

SP - 70

EP - 73

BT - First International IEEE Workshop on Spoken Language Technology SLT 2006

A2 - Raorke, B.

PB - IEEE Computer Society

CY - Palm Beach

ER -