A Statistical, Grammar-Based Approach to Micro-Planning

Abstract:

(Joint work with Laura Perez-Beltrachini)

While there has been much work in recent years on data-driven natural language generation, little attention has been paid to the fine grained interactions that arise during micro-planning between aggregation, surface realization and sentence segmentation. In this talk, I will argue for a hybrid symbolic/statistical approach to jointly model the interactions arising in Natural Language Generation between syntactic, aggregation and sentence segmentation choices. The approach integrates a small hand-written grammar, a statistical hypertagger and a surface realization algorithm. It is applied to the verbalization of knowledge base queries and tested on 13 knowledge bases to demonstrate domain independence. We evaluate our approach in several ways. A quantitative analysis shows that the hybrid approach outperforms a purely symbolic approach in terms of both speed and coverage. Results from a human study indicate that users find the output of this hybrid statistic/symbolic system more fluent than both a template- and a purely symbolic grammar-based approach. Finally, we illustrate by means of examples that our approach can account for various factors impacting aggregation, sentence segmentation and surface realization.

Bio:

Claire Gardent is a senior researcher at the French National Center for Scientific Research (CNRS). She graduated from the University of Edinburgh and has worked on the automatic acquisition of lexical resources for French, syntactic parsing, semantic role labelling, text generation and the interaction between virtual worlds and natural language processing. Her current research focuses on generating text from semantic web data.

Claire Gardent was nominated Chair of the European Chapter for the Association of Computational Linguistics (EACL), editor in chief of the journals "Traitement Automatique des Langues" and "Language and Linguistic Compass (Computational and Mathematical Section)" and member of the editorial board of the journals "Computational Linguistics", "Journal of Semantics", "Journal of Linguistic Modelling". She acted as program chair for various international conferences, workshops and summer schools (EACL, ENLG, SemDIAL, SIGDIAL, ESSLLI). She currently heads the WebNLG project (Nancy, Bolzano, Stanford SRI) and is the chair of SIGGEN, the ACL Special Interest Group in Natural Language Generation.