Loading [MathJax]/extensions/tex2jax.js

The Impact of Modeling Overall Argumentation with Tree Kernels

Publikation: Beitrag in Buch/Bericht/Sammelwerk/KonferenzbandAufsatz in KonferenzbandForschungPeer-Review

Autorschaft

Externe Organisationen

  • Bauhaus-Universität Weimar
  • Qatar Computing Research institute

Details

OriginalspracheEnglisch
Titel des SammelwerksProceedings of the 2017 Conference on Empirical Methods in Natural Language Processing
ErscheinungsortCopenhagen
Seiten2379-2389
Seitenumfang11
PublikationsstatusVeröffentlicht - Aug. 2017
Extern publiziertJa
Veranstaltung2017 Conference on Empirical Methods in Natural Language Processing - Copenhagen, Dänemark
Dauer: 7 Sept. 201711 Sept. 2017

Abstract

Several approaches have been proposed to model either the explicit sequential structure of an argumentative text or its implicit hierarchical structure. So far, the adequacy of these models of overall argumentation remains unclear. This paper asks what type of structure is actually important to tackle downstream tasks in computational argumentation. We analyze patterns in the overall argumentation of texts from three corpora. Then, we adapt the idea of positional tree kernels in order to capture sequential and hierarchical argumentative structure together for the first time. In systematic experiments for three text classification tasks, we find strong evidence for the impact of both types of structure. Our results suggest that either of them is necessary while their combination may be beneficial.

ASJC Scopus Sachgebiete

Zitieren

The Impact of Modeling Overall Argumentation with Tree Kernels. / Wachsmuth, Henning; da San Martino, Giovanni; Kiesel, Dora et al.
Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Copenhagen, 2017. S. 2379-2389.

Publikation: Beitrag in Buch/Bericht/Sammelwerk/KonferenzbandAufsatz in KonferenzbandForschungPeer-Review

Wachsmuth, H, da San Martino, G, Kiesel, D & Stein, B 2017, The Impact of Modeling Overall Argumentation with Tree Kernels. in Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Copenhagen, S. 2379-2389, 2017 Conference on Empirical Methods in Natural Language Processing, Copenhagen, Dänemark, 7 Sept. 2017. https://doi.org/10.18653/v1/d17-1253
Wachsmuth, H., da San Martino, G., Kiesel, D., & Stein, B. (2017). The Impact of Modeling Overall Argumentation with Tree Kernels. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing (S. 2379-2389). https://doi.org/10.18653/v1/d17-1253
Wachsmuth H, da San Martino G, Kiesel D, Stein B. The Impact of Modeling Overall Argumentation with Tree Kernels. in Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Copenhagen. 2017. S. 2379-2389 doi: 10.18653/v1/d17-1253
Wachsmuth, Henning ; da San Martino, Giovanni ; Kiesel, Dora et al. / The Impact of Modeling Overall Argumentation with Tree Kernels. Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Copenhagen, 2017. S. 2379-2389
Download
@inproceedings{02b87469a1df4ccf8229ab02532ac0ef,
title = "The Impact of Modeling Overall Argumentation with Tree Kernels",
abstract = "Several approaches have been proposed to model either the explicit sequential structure of an argumentative text or its implicit hierarchical structure. So far, the adequacy of these models of overall argumentation remains unclear. This paper asks what type of structure is actually important to tackle downstream tasks in computational argumentation. We analyze patterns in the overall argumentation of texts from three corpora. Then, we adapt the idea of positional tree kernels in order to capture sequential and hierarchical argumentative structure together for the first time. In systematic experiments for three text classification tasks, we find strong evidence for the impact of both types of structure. Our results suggest that either of them is necessary while their combination may be beneficial.",
author = "Henning Wachsmuth and {da San Martino}, Giovanni and Dora Kiesel and Benno Stein",
year = "2017",
month = aug,
doi = "10.18653/v1/d17-1253",
language = "English",
isbn = "9781945626838",
pages = "2379--2389",
booktitle = "Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing",
note = "2017 Conference on Empirical Methods in Natural Language Processing, EMNLP 2017 ; Conference date: 07-09-2017 Through 11-09-2017",

}

Download

TY - GEN

T1 - The Impact of Modeling Overall Argumentation with Tree Kernels

AU - Wachsmuth, Henning

AU - da San Martino, Giovanni

AU - Kiesel, Dora

AU - Stein, Benno

PY - 2017/8

Y1 - 2017/8

N2 - Several approaches have been proposed to model either the explicit sequential structure of an argumentative text or its implicit hierarchical structure. So far, the adequacy of these models of overall argumentation remains unclear. This paper asks what type of structure is actually important to tackle downstream tasks in computational argumentation. We analyze patterns in the overall argumentation of texts from three corpora. Then, we adapt the idea of positional tree kernels in order to capture sequential and hierarchical argumentative structure together for the first time. In systematic experiments for three text classification tasks, we find strong evidence for the impact of both types of structure. Our results suggest that either of them is necessary while their combination may be beneficial.

AB - Several approaches have been proposed to model either the explicit sequential structure of an argumentative text or its implicit hierarchical structure. So far, the adequacy of these models of overall argumentation remains unclear. This paper asks what type of structure is actually important to tackle downstream tasks in computational argumentation. We analyze patterns in the overall argumentation of texts from three corpora. Then, we adapt the idea of positional tree kernels in order to capture sequential and hierarchical argumentative structure together for the first time. In systematic experiments for three text classification tasks, we find strong evidence for the impact of both types of structure. Our results suggest that either of them is necessary while their combination may be beneficial.

UR - http://www.scopus.com/inward/record.url?scp=85061325760&partnerID=8YFLogxK

U2 - 10.18653/v1/d17-1253

DO - 10.18653/v1/d17-1253

M3 - Conference contribution

AN - SCOPUS:85061325760

SN - 9781945626838

SP - 2379

EP - 2389

BT - Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing

CY - Copenhagen

T2 - 2017 Conference on Empirical Methods in Natural Language Processing, EMNLP 2017

Y2 - 7 September 2017 through 11 September 2017

ER -

Von denselben Autoren