Section: New Results

Large scale corpus processing

Participant : Éric Villemonte de La Clergerie.

In the context of the PASSAGE action, we have continued the explore the use of distributed computing for processing of large corpora, largely using GRID 5000 and a local cluster at INRIA Rocquencourt. We use more and more such resources also for the post-parsing phases and the ambition is to use them for machine-learning phases.

GRID5000 and the local cluster were specially useful for the parsing evaluation campaign (October and November 2009), even such real life experiments tend to show that scripts in such complex environments are never robust enough.