Enrichment of Renaissance texts with proper names

Denis Maurel (1) , Nathalie Friburger (1) , Iris Eshkol-Taravella (2)

1 BDTLN - Bases de données et traitement des langues naturelles
2 LLL - Laboratoire Ligérien de Linguistique

The Renom project proposes to enrich Renaissance texts by proper names. These texts present two new challenges: great diversity due to various spellings of words; numerous XML-TEI tags to save the exact format of original edition. The task consisted to add Named Entity tags to this format tagging with generally the left context and sometimes the right context of a name. To do that, we improved the free and open source program CasSys to parse texts with Unitex graph cascades and we built dictionaries and specific cascades. The slot error rate was 6.1%. Proper Names and maps. were to allow navigating into. So, this paper deals with Named Entity Recognition in Renaissance texts.

Mots clés: Named entities; Renaissance texts; Graph cascades; CasSys; Humanities and tourism

Domaines: Linguistique; Informatique [cs]

Denis Maurel, Nathalie Friburger, Iris Eshkol-Taravella. Enrichment of Renaissance texts with proper names. INFOtheca : Journal of Information and Library Science, 2014, 15 (1), pp.15-27. ⟨hal-01174733⟩