EmcSci

1er atelier sur l' Extraction et la Modélisation de Connaissances à partir de textes scientifiques

IC 2017, Caen, France

Programme

Lundi 3 juillet 2017

Campus 2 de l'université de Caen (salle à préciser)

  • 14:00-14:10 Bienvenue et présentation de l'atelier (D.Buscaldi)
  • 14:10-15:10 Conférence invitée de Francesco Osborne:

"Research 3.0: Integrating Knowledge Graphs in the Research Process"

The number of papers and the available scientific knowledge is growing rapidly, making increasingly harder to keep track of all the relevant knowledge that could facilitate a research endeavour. The availability of research materials on the Web and the presence of systems for exploring the research environment alleviate this issue only to some degree. I suggest that we need to switch to a modern research paradigm in which researchers will be assisted by software capable of applying data-driven methodologies on a machine-readable description of research knowledge. The aim is expanding the conceptual horizon of researchers and combining human creativity with the data mining ability of computers. I will elaborate on this perspective and discuss relevant research trends.

  • 15:10-15:30 Unsupervised relation extraction from scientific texts using self-organizing maps (Elena Manishina, Mouna Kamel, Cassia Trojahn and Nathalie Aussenac-Gilles)
  • 15:30-16:00 Pause
  • 16:00-16:20 Part-of-Speech Tag Sequences to Filter Candidates for Keyphrase Extraction from Scientific Publications (Simon David Hernandez, Davide Buscaldi and Thierry Charnois)
  • 16:20-16:40 A propos des territoires dans les corpus scientifiques (Marie-Noelle Bessagnet, Eric Kergosien, Mohammadamin Farvardin, Annig Le Parc - Lacayrelle and Christian Sallaberry)
  • 16:40-17:00 Discussion et conclusion