Paris NLP First Meetup

Seating is on a first come, first served basis whether you have RSVPed or not, so we suggest arriving early. We can host 150 people.

La salle permet d'accueillir 150 personnes. L'inscription est obligatoire mais ne garantit pas que vous pourrez entrer, nous vous recommandons donc d'arriver un peu en avance.

• Nicolas Marlier, Co-Founder & CTO at JulieDesk (https://www.juliedesk.com/).

Building a supervised AI system that schedules meetings. A few words about Julie Desk (https://www.juliedesk.com/), and how it works. How we designed our hybrid AI-human system, a few algorithms explained.

• Djamé Seddah, tenured associate professor in linguistics and informatics at the University Paris Sorbonne (Paris 4) and member of the INRIA’s Alpage project. Djamé explores the impact of annotation schemes of non-canonical languages for French as well as out of domain parsing, noisy user generated content parsing and syntax- semantic interface. Djamé is one of the founders of the statistical parsing of morphologically-rich languages community and has instigated the creation of many annotated data set for French (the Sequoia Treebank, the French Social Media Bank and the French Question Bank).

Because Syntax Does Matter: Improving Predicate-Argument Structures Parsing with Syntactic Features, The Case of French.

Parsing predicate-argument structures in a deep syntax framework requires graphs to be predicted. Argument structures represent a higher level of abstraction than the syntactic ones and are thus more difficult to predict even for highly accurate parsing models on surfacic syntax. In this talk we investigate deep syntax parsing, using a French data set (Ribeyre et al., 2014a). We demonstrate that the use of topologically different types of syntactic features, such as dependencies, tree fragments, spines or syntactic paths, brings a much needed context to the parser. Our higher-order parsing model, gaining thus up to 4 points, establishes the state of the art for parsing French deep syntactic structures. This talk presents a work that was carried out during Corentin Ribeyre's PhD thesis, jointly supervised with Éric de la Clergerie.

• Claude Fauconnet, Digital Innovation Project Manager at Total Holding. Claude has 20 years of experience in oil exploration in the Total Group. He now evaluates, contextualizes and disseminates new digital technologies in the different business lines. His practice focuses on semantic technologies, Big Data, and on issues related to the application field.

Claude will talk about the various applications of Text Mining at Total.