Adaptation of a named entity recognition system for the ESTER 2 evaluation campaign

In this paper, we report our participation to the ESTER 2 (Evaluation des Systemes de Transcription Enrichie d'Emissions Radiophoniques) evaluation campaign, on the Named Entity Recognition for French track. After describing the ESTER 2 goals and guidelines, we present our deep robust parser. T...

Full description

Saved in:
Bibliographic Details
Main Authors: Brun, C., Ehrmann, M.
Format: Conference Proceeding
Language:English
Subjects:
Data mining
Europe
Evaluation
Event detection
Guidelines
Information Extraction
Named Entity Detection
Natural language processing
Natural languages
Radio broadcasting
Robustness
Speech processing
Testing
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we report our participation to the ESTER 2 (Evaluation des Systemes de Transcription Enrichie d'Emissions Radiophoniques) evaluation campaign, on the Named Entity Recognition for French track. After describing the ESTER 2 goals and guidelines, we present our deep robust parser. Then we show how we adapt the existing French NER module of this parser to the ESTER 2 task. The results we obtain both on manual and automatic transcriptions are presented and discussed.
DOI:10.1109/NLPKE.2009.5313775