Bois, Remi, Leveling, Johannes ORCID: 0000-0003-0603-4191, Goeuriot, Lorraine ORCID: 0000-0001-7491-1980, Jones, Gareth J.F. ORCID: 0000-0002-4033-9135 and Kelly, Liadh ORCID: 0000-0003-1131-5238 (2014) Porting a summarizer to the French language. In: Traitement Automatique du Langage Naturel (TALN), 1-4 Jul 2014, Marseille, France..
Abstract
We describe the porting of the English language REZIME text summarizer to the French language. REZIME
is a single-document summarizer particularly focused on summarization of medical documents. Summaries are
created by extracting key sentences from the original document. The sentence selection employs machine learning techniques,
using statistical, syntactic and lexical features which are computed based on specialized language resources. The
REZIME system was initially developed for English documents.In this paper we present the summarizer architecture, and
describe the steps required to adapt it to the French language. The summarizer performance is evaluated for English and
French datasets. Results show that the adaptation to French results in a system performance comparable to English.
Metadata
Item Type: | Conference or Workshop Item (Poster) |
---|---|
Event Type: | Conference |
Refereed: | Yes |
Uncontrolled Keywords: | Single-document summarization; Multilingual; Medical domain |
Subjects: | Computer Science > Information retrieval |
DCU Faculties and Centres: | Research Initiatives and Centres > Centre for Next Generation Localisation (CNGL) DCU Faculties and Schools > Faculty of Engineering and Computing > School of Computing |
Use License: | This item is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 3.0 License. View License |
ID Code: | 20114 |
Deposited On: | 17 Sep 2014 09:27 by Liadh Kelly . Last Modified 25 Oct 2018 09:31 |
Documents
Full text available as:
Preview |
PDF
- Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
325kB |
Downloads
Downloads
Downloads per month over past year
Archive Staff Only: edit this record