Improving the quality of minority class identification in dialog act tagging

Adinoyi Omuya; Vinodkumar Prabhakaran; Owen Rambow

Improving the quality of minority class identification in dialog act tagging

Adinoyi Omuya
, Vinodkumar Prabhakaran
, Owen Rambow

Linguistics

10gen
Columbia University

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

11 Scopus citations

Abstract

We present a method of improving the performance of dialog act tagging in identifying minority classes by using per-class feature optimization and a method of choosing the class based not on confidence, but on a cascade of classifiers. We show that it gives a minority class F-measure error reduction of 22.8%, while also reducing the error for other classes and the overall error by about 10%.

Original language	English
Title of host publication	Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics
Subtitle of host publication	Human Language Technologies
Publisher	Association for Computational Linguistics (ACL)
Pages	802-807
Number of pages	6
ISBN (Electronic)	9781937284473
State	Published - 2013
Event	2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2013 - Atlanta, United States Duration: Jun 9 2013 → Jun 14 2013

Publication series

Name	NAACL HLT 2013 - 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Main Conference

Conference

Conference	2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2013
Country/Territory	United States
City	Atlanta
Period	06/9/13 → 06/14/13

Cite this

Omuya, A., Prabhakaran, V., & Rambow, O. (2013). Improving the quality of minority class identification in dialog act tagging. In Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (pp. 802-807). (NAACL HLT 2013 - 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Main Conference). Association for Computational Linguistics (ACL).

Omuya, Adinoyi ; Prabhakaran, Vinodkumar ; Rambow, Owen. / Improving the quality of minority class identification in dialog act tagging. Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Association for Computational Linguistics (ACL), 2013. pp. 802-807 (NAACL HLT 2013 - 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Main Conference).

@inproceedings{0fc10c8a39c34c7f8369eacd7c1c9397,

title = "Improving the quality of minority class identification in dialog act tagging",

abstract = "We present a method of improving the performance of dialog act tagging in identifying minority classes by using per-class feature optimization and a method of choosing the class based not on confidence, but on a cascade of classifiers. We show that it gives a minority class F-measure error reduction of 22.8\%, while also reducing the error for other classes and the overall error by about 10\%.",

author = "Adinoyi Omuya and Vinodkumar Prabhakaran and Owen Rambow",

note = "Publisher Copyright: {\textcopyright} 2013 Association for Computational Linguistics.; 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2013 ; Conference date: 09-06-2013 Through 14-06-2013",

year = "2013",

language = "English",

series = "NAACL HLT 2013 - 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Main Conference",

publisher = "Association for Computational Linguistics (ACL)",

pages = "802--807",

booktitle = "Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics",

}

Omuya, A, Prabhakaran, V & Rambow, O 2013, Improving the quality of minority class identification in dialog act tagging. in Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. NAACL HLT 2013 - 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Main Conference, Association for Computational Linguistics (ACL), pp. 802-807, 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2013, Atlanta, United States, 06/9/13.

Improving the quality of minority class identification in dialog act tagging. / Omuya, Adinoyi; Prabhakaran, Vinodkumar; Rambow, Owen.
Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Association for Computational Linguistics (ACL), 2013. p. 802-807 (NAACL HLT 2013 - 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Main Conference).

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

TY - GEN

T1 - Improving the quality of minority class identification in dialog act tagging

AU - Omuya, Adinoyi

AU - Prabhakaran, Vinodkumar

AU - Rambow, Owen

PY - 2013

Y1 - 2013

N2 - We present a method of improving the performance of dialog act tagging in identifying minority classes by using per-class feature optimization and a method of choosing the class based not on confidence, but on a cascade of classifiers. We show that it gives a minority class F-measure error reduction of 22.8%, while also reducing the error for other classes and the overall error by about 10%.

AB - We present a method of improving the performance of dialog act tagging in identifying minority classes by using per-class feature optimization and a method of choosing the class based not on confidence, but on a cascade of classifiers. We show that it gives a minority class F-measure error reduction of 22.8%, while also reducing the error for other classes and the overall error by about 10%.

UR - https://www.scopus.com/pages/publications/84906925903

M3 - Conference contribution

AN - SCOPUS:84906925903

T3 - NAACL HLT 2013 - 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Main Conference

SP - 802

EP - 807

BT - Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics

PB - Association for Computational Linguistics (ACL)

T2 - 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2013

Y2 - 9 June 2013 through 14 June 2013

ER -

Omuya A, Prabhakaran V, Rambow O. Improving the quality of minority class identification in dialog act tagging. In Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Association for Computational Linguistics (ACL). 2013. p. 802-807. (NAACL HLT 2013 - 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Main Conference).

Improving the quality of minority class identification in dialog act tagging

Abstract

Publication series

Conference

Other files and links

Fingerprint

Cite this