TALN - Automatic Natural Language
The team's work focuses on written language analysis and falls under two main research themes :
Analysis & Discovery
The analysis traditionally focuses on formal models of language syntax and semantics. We work on lexicalized grammars allowing a syntactic analysis in dependence and on probabilistic grammars. The discovery applies various methods of analysis to textual data corpuses to isolate remarkable elements. The team has a strong expertise in the processing of documents belonging to specialized fields.
Alignment & Multilingualism
In this theme, we study methods for reconciling various data sources to obtain additional information: alignments. We work on the alignment of comparable corpora, texts in two languages without translation reports, multimodal corpora, texts from oral or handwritten sources and written texts.