Skip to Main content Skip to Navigation
Theses

Apprentissage de connaissances morphologiques pour l'acquisition automatique de ressources lexicales

Delphine Bernhard 1
1 OSIRIS
TIMC - Techniques de l'Ingénierie Médicale et de la Complexité - Informatique, Mathématiques et Applications, Grenoble - UMR 5525
Abstract : Lexico-semantic resources, like thesauri, terminologies and ontologies, aim at organising knowledge by detailing semantic relationships such as synonymy or specialisation. The cost for manually building this kind of resources is high. Methods for the automatic acquisition of knowledge from text corpora are therefore widely used. These methods aim at automatically extracting terms and semantic relationships. In this thesis, we investigate the role which can be played by morphology, i.e. the internal structure of words, within such systems.

We describe two methods for the unsupervised acquisition of morphological knowledge. The first one segments words into sub-units while the other conflates words in morphological families.

We then explore possible uses for this kind of knowledge. We re-use morphological families to weight and visualise keywords. We also define patterns based on morphological segmentation which make it possible to discover semantic relationships such as hypernymy and co-hyponymy.
Complete list of metadata

https://tel.archives-ouvertes.fr/tel-00119257
Contributor : Delphine Bernhard <>
Submitted on : Friday, December 8, 2006 - 12:44:34 PM
Last modification on : Thursday, November 19, 2020 - 3:56:08 PM
Long-term archiving on: : Thursday, September 20, 2012 - 3:41:01 PM

Identifiers

  • HAL Id : tel-00119257, version 1

Collections

Citation

Delphine Bernhard. Apprentissage de connaissances morphologiques pour l'acquisition automatique de ressources lexicales. Autre [cs.OH]. Université Joseph-Fourier - Grenoble I, 2006. Français. ⟨tel-00119257⟩

Share

Metrics

Record views

919

Files downloads

2304