Concept Discovery for Language Understanding in an Information-Query Dialogue System

Abstract : Most recent efficient statistical approaches for natural language understanding require a segmental annotation of training data. Such an annotation implies both to determine the concepts in a sentence and to link them to their corresponding word segments. In this paper we propose a two-steps alternative to the fully manual annotation of data: an initial unsupervised concept discovery, based on latent Dirichlet allocation, is followed by an automatic segmentation using integer linear optimisation. The relation between discovered topics and task-dependent concepts is evaluated on a spoken dialogue task for which a reference annotation is available. Topics and concepts are shown close enough to achieve a potential reduction of one half of the manual annotation cost.
Complete list of metadatas

Cited literature [10 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01314539
Contributor : Bibliothèque Universitaire Déposants Hal-Avignon <>
Submitted on : Wednesday, February 27, 2019 - 10:08:19 AM
Last modification on : Wednesday, May 15, 2019 - 10:12:03 AM
Long-term archiving on : Tuesday, May 28, 2019 - 12:58:58 PM

File

KDIR11.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01314539, version 1

Collections

Citation

Nathalie Camelin, Boris Detienne, Stéphane Huet, Dominique Quadri, Fabrice Lefèvre. Concept Discovery for Language Understanding in an Information-Query Dialogue System. International Conference on Knowledge Discovery and Information Retrieval (KDIR), Oct 2011, Paris, France. ⟨hal-01314539⟩

Share

Metrics

Record views

66

Files downloads

39