Section: Software and Platforms

MElt

Participants : Benoît Sagot [correspondant] , Pierre Magistry.

MElt is a part-of-speech tagger, initially developed in collaboration with Pascal Denis (Magnet, Inria — then at Alpage), which was trained for French (on the French TreeBank and coupled with the Lefff), also trained on English [63] , Spanish [69] , Italian [94] , German, Dutch, Polish, Kurmanji Kurdish [104] and Persian [89] , [90] . It is state-of-the-art for French.

It is now able to handle noisy corpora (French and English only).

MElt also includes a lemmatization post-processing step.

A specific effort has been made towards the usability of MElt by linguists. In particular, a training session has been organized, and a user guide has been written.

Moreover, a preliminary version of MElt which accepts input DAGs has been developed.

MElt is distributed freely as a part of the Alpage linguistic workbench.