The CEA LIST is building a large semantic map of the French language. Web pages are downloaded and submittted to a language identifier and then French pages are sent to a syntactic analyzer. All syntactic dependency relations for all words are extracted an stored in a custom database built by the CEA LIST. This demo shows a user interface into this database, and shows the type of relations that are extracted for each word. This complete map of the French language has applications in classification, speech recognition, machine translation, and indexing. Part of the funding for this research comes from the Jean-Luc Lagardere Foundation.

Loading more stuff…

Hmm…it looks like things are taking a while to load. Try again?

Loading videos…