FENEC : un corpus équilibré pour l'évaluation des entités nommées en français (FENEC : a balanced sample corpus for French named entity recognition ).
Alice MillourYoann DupontAlexane JouglarKarën FortPublished in: TALN-RECITAL (2022)
Keyphrases
- named entity recognition
- annotated corpus
- information extraction
- named entities
- reference resolution
- linguistic features
- natural language processing
- named entity disambiguation
- genia corpus
- maximum entropy
- text summarization
- relation extraction
- semi supervised
- automatic annotation
- text mining
- conditional random fields
- maximum likelihood
- proper names
- knn
- object recognition