Paper Info
Title
Empirically-based control of natural language generation
Abstract
In this paper we present a new approach to controlling the behaviour of a natural language generation system by correlating internal decisions taken during free generation of a wide range of texts with the surface stylistic characteristics of the resulting outputs, and using the correlation to control the generator. This contrasts with the generate-and-test architecture adopted by most previous empirically-based generation approaches, offering a more efficient, generic and holistic method of generator control. We illustrate the approach by describing a system in which stylistic variation (in the sense of Biber (1988)) can be effectively controlled during the generation of short medical information texts.
Year
DOI
Venue
2005
10.3115/1219840.1219848
ACL
Keywords
Field
DocType
natural language generation system,generator control,previous empirically-based generation approach,holistic method,surface stylistic characteristic,empirically-based control,internal decision,stylistic variation,new approach,generate-and-test architecture,free generation
Natural language generation,Architecture,Computer science,Contrast (statistics),Artificial intelligence,Natural language processing,Machine learning,Generator control
Conference
Volume
Citations 
PageRank 
P05-1
22
1.18
References 
Authors
5
2
Name
Order
Citations
PageRank
Daniel S. Paiva1372.95
Roger Evans234455.12