Semi-supervised SRL system with Bayesian inference - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2014

Semi-supervised SRL system with Bayesian inference

Résumé

We propose a new approach to perform semi-supervised training of Semantic Role Labeling models with very few amount of initial labeled data. The proposed approach combines in a novel way supervised and unsupervised training, by forcing the supervised classifier to over-generate potential semantic candidates, and then letting unsupervised inference choose the best ones. Hence, the supervised classifier can be trained on a very small corpus and with coarse-grain features, because its precision does not need to be high: its role is mainly to constrain Bayesian inference to explore only a limited part of the full search space. This approach is evaluated on French and English. In both cases, it achieves very good performance and outperforms a strong supervised baseline when only a small number of annotated sentences is available and even without using any previously trained syntactic parser.
Fichier principal
Vignette du fichier
semisup.pdf (147.1 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01015414 , version 1 (26-06-2014)

Identifiants

  • HAL Id : hal-01015414 , version 1

Citer

Alejandra Lorenzo, Christophe Cerisara. Semi-supervised SRL system with Bayesian inference. 15th International Conference on Intelligent Text Processing and Computational Linguistics, Apr 2014, Kathmandu, Nepal. pp.433. ⟨hal-01015414⟩
175 Consultations
199 Téléchargements

Partager

Gmail Facebook X LinkedIn More