DaFOE : an Ontology Building Platform From Texts or Thesauri?

نویسندگان

  • Jean Charlet
  • Sylvie Szulman
  • Nathalie Aussenac-Gilles
  • Adeline Nazarenko
  • Nathalie Hernandez
  • Nadia Nadah
  • Eric Sardet
  • Jean Delahousse
  • Henry Valéry Téguiak
چکیده

La construction d'ontologies à partir de textes fait l'objet d'études depuis plusieurs années dans le domaine de l'ingénierie des ontologies (Terminae Aussenac-Gilles et al. (2000), Text2Onto Cimiano & Volker (2005)). Un cadre méthodologique en quatre étapes (constitution d'un corpus de documents, analyse linguistique du corpus, conceptualisation, opérationnalisation de l'ontologie) est commun à la plupart des méthodes de construction d'ontologies à partir de textes. S'il existe plusieurs plateformes de traitement automatique de la langue (TAL) permettant d'analyser automatiquement les corpus et de les annoter tant du point de vue syntaxique que statistique, il n'existe actuellement aucune procédure généralement acceptée, ni a fortiori aucun ensemble cohérent d'outils supports, permettant de concevoir de façon progressive, explicite et traçable une ontologie de domaine à partir d'un ensemble de ressources informationnelles relevant de ce domaine. C'est ce que nous proposons dans la plateforme Dafoe, en même temps que nous cherchons à favoriser l'émergence de tels outils se développant comme gre on.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

DAFOE: a Platform for Building Ontologies from Texts

Although text-based ontology engineering gained much popularity in the last 10 years, very few ontology engineering platforms exploit the full potential of the connection between texts and ontologies. We propose DAFOE, a new platform for building ontologies with a terminological component using different types of linguistic entries (text corpora, results of natural language processing tools, te...

متن کامل

Incremental Design of Ontologies - A Model Transformation-based Approach

This paper focuses on Model Driven Engineering (MDE) techniques to build ontology incrementally. The global building process consists of building ontologies from texts through a stepwise approach. Our approach consists of modeling each step independently and defining a mapping model to express relationships between these steps. Results of this work are applied in the ANR DAFOE project, which ai...

متن کامل

DAFOE: A Multimodel and Multimethod Platform for Building Domain Ontologies∗

The concept of ontologies, appeared in the nineties, constitute a key point to represent and share the meaning carried by formal symbols. Yet building of such an ontology is quite difficult. A way to do so is to use preexistent elements (textual corpus, taxonomies, norms or other ontologies) and operate them as a basis to define the field ontology. Still, there is neither accepted process nor s...

متن کامل

Building Medical Ontologies Based on Terminology Extraction from Texts: an Experimentation in Pneumology

Pathologies and acts are classified in thesauri to help physicians to code their activity. In practice, the use of thesauri is not sufficient to reduce variability in coding and thesauri do not fit computer processing. We think the automation of the coding task requires a conceptual modelling of medical items: an ontology. Our objective is to help pneumologists code acts and diagnoses with a so...

متن کامل

Pruning-based Identification of Domain Ontologies

We present a novel approach of extracting a domain ontology from large-scale thesauri. Concepts are identified to be relevant for a domain based on their frequent occurrence in domain texts. The approach allows to bootstrap the ontology engineering process from given legacy thesauri and identifies an initial domain ontology that may easily be refined by experts in a later stage. We present a th...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2009