DEBACER: a method for slicing moderated debates

Thomas Palmeira Ferraz; Alexandre Alcoforado; Enzo Bustos; André Seidel Oliveira; Rodrigo Gerber; Naíde Müller; André Corrêa d’Almeida; Bruno Miguel Veloso; Anna Helena Reali Costa

doi:10.5753/eniac.2021.18293

DEBACER: a method for slicing moderated debates

Thomas Palmeira Ferraz, Alexandre Alcoforado, Enzo Bustos, André Seidel Oliveira, Rodrigo Gerber, Naíde Müller, André Corrêa d’Almeida, Bruno Miguel Veloso, Anna Helena Reali Costa

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

18 Downloads

Abstract

Subjects change frequently in moderated debates with several participants, such as in parliamentary sessions, electoral debates, and trials. Partitioning a debate into blocks with the same subject is essential for understanding. Often a moderator is responsible for defining when a new block begins so that the task of automatically partitioning a moderated debate can focus solely on the moderator's behavior. In this paper, we (i) propose a new algorithm, DEBACER, which partitions moderated debates; (ii) carry out a comparative study between conventional and BERTimbau pipelines; and (iii) validate DEBACER applying it to the minutes of the Assembly of the Republic of Portugal. Our results show the effectiveness of DEBACER.

Original language	English
Title of host publication	ENIAC 2021
Subtitle of host publication	XVIII encontro nacional de inteligência artificial e computacional. 18th national meeting on artificial and computational intelligence
Editors	Elaine Ribeiro de Faria, Paulo de Tarso Guerra
Publisher	Sociedade Brasileira de Computação
Pages	667-678
Number of pages	12
DOIs	https://doi.org/10.5753/eniac.2021.18293
Publication status	Published - 29 Nov 2021

Keywords

Language processing
Political documents
Spoken text processing
Speech split
Dialogue partitioning

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.5753/eniac.2021.18293Licence: Unspecified

42099194Final published version, 416 KBLicence: Unspecified

http://hdl.handle.net/10400.14/37105Licence: Unspecified

Cite this

Ferraz, T. P., Alcoforado, A., Bustos, E., Oliveira, A. S., Gerber, R., Müller, N., d’Almeida, A. C., Veloso, B. M., & Costa, A. H. R. (2021). DEBACER: a method for slicing moderated debates. In E. R. D. Faria, & P. D. T. Guerra (Eds.), ENIAC 2021: XVIII encontro nacional de inteligência artificial e computacional. 18th national meeting on artificial and computational intelligence (pp. 667-678). Sociedade Brasileira de Computação. https://doi.org/10.5753/eniac.2021.18293

Ferraz, Thomas Palmeira ; Alcoforado, Alexandre ; Bustos, Enzo et al. / DEBACER : a method for slicing moderated debates. ENIAC 2021: XVIII encontro nacional de inteligência artificial e computacional. 18th national meeting on artificial and computational intelligence. editor / Elaine Ribeiro de Faria ; Paulo de Tarso Guerra. Sociedade Brasileira de Computação, 2021. pp. 667-678

@inproceedings{2e16d8ba1bd14ce4b83efac1903c214e,

title = "DEBACER: a method for slicing moderated debates",

abstract = "Subjects change frequently in moderated debates with several participants, such as in parliamentary sessions, electoral debates, and trials. Partitioning a debate into blocks with the same subject is essential for understanding. Often a moderator is responsible for defining when a new block begins so that the task of automatically partitioning a moderated debate can focus solely on the moderator's behavior. In this paper, we (i) propose a new algorithm, DEBACER, which partitions moderated debates; (ii) carry out a comparative study between conventional and BERTimbau pipelines; and (iii) validate DEBACER applying it to the minutes of the Assembly of the Republic of Portugal. Our results show the effectiveness of DEBACER.",

keywords = "Language processing, Political documents, Spoken text processing, Speech split, Dialogue partitioning",

author = "Ferraz, {Thomas Palmeira} and Alexandre Alcoforado and Enzo Bustos and Oliveira, {Andr{\'e} Seidel} and Rodrigo Gerber and Na{\'i}de M{\"u}ller and d{\textquoteright}Almeida, {Andr{\'e} Corr{\^e}a} and Veloso, {Bruno Miguel} and Costa, {Anna Helena Reali}",

year = "2021",

month = nov,

day = "29",

doi = "10.5753/eniac.2021.18293",

language = "English",

pages = "667--678",

editor = "Faria, {Elaine Ribeiro de} and Guerra, {Paulo de Tarso}",

booktitle = "ENIAC 2021",

publisher = "Sociedade Brasileira de Computa{\c c}{\~a}o",

address = "Brazil",

}

Ferraz, TP, Alcoforado, A, Bustos, E, Oliveira, AS, Gerber, R, Müller, N, d’Almeida, AC, Veloso, BM & Costa, AHR 2021, DEBACER: a method for slicing moderated debates. in ERD Faria & PDT Guerra (eds), ENIAC 2021: XVIII encontro nacional de inteligência artificial e computacional. 18th national meeting on artificial and computational intelligence. Sociedade Brasileira de Computação, pp. 667-678. https://doi.org/10.5753/eniac.2021.18293

DEBACER: a method for slicing moderated debates. / Ferraz, Thomas Palmeira; Alcoforado, Alexandre; Bustos, Enzo et al.
ENIAC 2021: XVIII encontro nacional de inteligência artificial e computacional. 18th national meeting on artificial and computational intelligence. ed. / Elaine Ribeiro de Faria; Paulo de Tarso Guerra. Sociedade Brasileira de Computação, 2021. p. 667-678.

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

TY - GEN

T1 - DEBACER

T2 - a method for slicing moderated debates

AU - Ferraz, Thomas Palmeira

AU - Alcoforado, Alexandre

AU - Bustos, Enzo

AU - Oliveira, André Seidel

AU - Gerber, Rodrigo

AU - Müller, Naíde

AU - d’Almeida, André Corrêa

AU - Veloso, Bruno Miguel

AU - Costa, Anna Helena Reali

PY - 2021/11/29

Y1 - 2021/11/29

N2 - Subjects change frequently in moderated debates with several participants, such as in parliamentary sessions, electoral debates, and trials. Partitioning a debate into blocks with the same subject is essential for understanding. Often a moderator is responsible for defining when a new block begins so that the task of automatically partitioning a moderated debate can focus solely on the moderator's behavior. In this paper, we (i) propose a new algorithm, DEBACER, which partitions moderated debates; (ii) carry out a comparative study between conventional and BERTimbau pipelines; and (iii) validate DEBACER applying it to the minutes of the Assembly of the Republic of Portugal. Our results show the effectiveness of DEBACER.

AB - Subjects change frequently in moderated debates with several participants, such as in parliamentary sessions, electoral debates, and trials. Partitioning a debate into blocks with the same subject is essential for understanding. Often a moderator is responsible for defining when a new block begins so that the task of automatically partitioning a moderated debate can focus solely on the moderator's behavior. In this paper, we (i) propose a new algorithm, DEBACER, which partitions moderated debates; (ii) carry out a comparative study between conventional and BERTimbau pipelines; and (iii) validate DEBACER applying it to the minutes of the Assembly of the Republic of Portugal. Our results show the effectiveness of DEBACER.

KW - Language processing

KW - Political documents

KW - Spoken text processing

KW - Speech split

KW - Dialogue partitioning

U2 - 10.5753/eniac.2021.18293

DO - 10.5753/eniac.2021.18293

M3 - Conference contribution

SP - 667

EP - 678

BT - ENIAC 2021

A2 - Faria, Elaine Ribeiro de

A2 - Guerra, Paulo de Tarso

PB - Sociedade Brasileira de Computação

ER -

Ferraz TP, Alcoforado A, Bustos E, Oliveira AS, Gerber R, Müller N et al. DEBACER: a method for slicing moderated debates. In Faria ERD, Guerra PDT, editors, ENIAC 2021: XVIII encontro nacional de inteligência artificial e computacional. 18th national meeting on artificial and computational intelligence. Sociedade Brasileira de Computação. 2021. p. 667-678 doi: 10.5753/eniac.2021.18293

DEBACER: a method for slicing moderated debates

Abstract

Keywords

UN SDGs

Access to Document

Fingerprint

Cite this