Surface Realisation from Knowledge-Bases

Bikash Gyawali (1) , Claire Gardent (1)

1 SYNALP - Natural Language Processing : representations, inference and semantics

We present a simple, data-driven approach to generation from knowledge bases (KB). A key feature of this approach is that grammar induction is driven by the extended domain of locality principle of TAG (Tree Adjoining Grammar); and that it takes into account both syntactic and semantic information. The resulting extracted TAG includes a unification based semantics and can be used by an existing surface realiser to generate sentences from KB data. Experimental evaluation on the KBGen data shows that our model outperforms a data-driven generate-and-rank approach based on an automatically induced probabilistic grammar; and is comparable with a handcrafted symbolic approach.

Mots clés: Natural Language Generation; Knowledge Bases; Computational Grammar; Grammar Induction

Domaines: Traitement du texte et du document

Bikash Gyawali, Claire Gardent. Surface Realisation from Knowledge-Bases. the 52nd Annual Meeting of the Association for Computational Linguistics, Jun 2014, Baltimore, United States. pp.424-434. ⟨hal-01021916⟩