Improving argument overlap for proposition-based summarisationReportar como inadecuado




Improving argument overlap for proposition-based summarisation - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

Publication Date: 2016-01-01

Journal Title: 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016 - Short Papers

Conference Name: Annual Meeting of the Association for Computational Linguists

Publisher: Association for Computational Linguistics

Pages: 479-485

Language: English

Type: Conference Object

This Version: VoR

Metadata: Show full item record

Citation: Fang, Y., & Teufel, S. (2016). Improving argument overlap for proposition-based summarisation. 54th Annual Meeting of the Association for Computational Linguistics, ACL 2016 - Short Papers, 479-485. http://anthology.aclweb.org/P16-2078

Abstract: We present improvements to our incremental proposition-based summariser, which is inspired by Kintsch and van Dijk's (1978) text comprehension model. Argument overlap is a central concept in this summariser. Our new model replaces the old overlap method based on distributional similarity with one based on lexical chains. We evaluate on a new corpus of 124 summaries of educational texts, and show that our new system outperforms the old method and several stateof-the-art non-proposition-based summarisers. The experiment also verifies that the incremental nature of memory cycles is beneficial in itself, by comparing it to a non-incremental algorithm using the same underlying information.

Sponsorship: The CSC Cambridge International Scholarship for the first author is gratefully acknowledged.

Identifiers:

External link: http://anthology.aclweb.org/P16-2078

This record's URL: https://www.repository.cam.ac.uk/handle/1810/267185

Rights: Attribution 4.0 International

Licence URL: http://creativecommons.org/licenses/by/4.0/





Autor: Fang, YTeufel, S

Fuente: https://www.repository.cam.ac.uk/handle/1810/267185



DESCARGAR PDF




Documentos relacionados