Paper Info
Title
Enabling Resource Sharing in Language Generation: an Abstract Reference Architecture.
Abstract
The RAGS project aims to develop a reference architecture for natural language generation, to facilitate modular devel opment of NLG systams as well as evaluation of components, systems and algorithms. This paper gives an overview of the proposed framework, describ- ing an abstract data model with five levels of representation : Conceptual, Semantic, Rhetorical, Document and Syntactic. We report on a re-implementation of an existing system using the RAGS data model. 1. The RAGS enterprise The primary goal of the RAGS project (Cahill et al., 1999a) is to develop a 'reference architecture' for applied natural language generation (NLG) systems. The aim is to produce an architectural specification which reflects mainstream current practice and provides a framework for the development of new applications and new components within NLG systems. The architecture is also intended to facilitate evaluation of NLG components, algorithms and systems. To achieve these goals, such an architecture has to be sufficiently conventional to be relevant to developers of existing systems, but also sufficiently generic and detailed to be useful as a resource for novel approaches. One of the distinctive properties of natural language generation when compared with other language engineer- ing applications is that it has to take seriously the full range of linguistic representation, from concepts through to mor- phology, or even phonetics. Any processing system is only as sophisticated as its input allows, so while a natural lan- guage understanding system might be judged by its syntac- tic prowess - even if its attention to semantics, pragmatics and underlying conceptual analysis is minimal - a genera- tion system is only as good as its deepest linguistic repre- sentations. This has particular implications for evaluation of NLG systems: it is hard to think of evaluation exercises along the lines of the MUC tasks (?) for information ex- traction, for instance, where the inputs consist of naturally- occurring text and the outputs are precisely specified by the evaluators. There is no general agreement on how inputs for NLG systems should be specified, nor on how output texts can be evaluated.
Year
Venue
Field
2000
LREC
Applications architecture,Programming language,Software engineering,Computer science,Natural language processing,Artificial intelligence,Reference architecture,Shared resource
DocType
Citations 
PageRank 
Conference
6
0.62
References 
Authors
6
9
Name
Order
Citations
PageRank
Lynne J. Cahill1357.54
Christy Doran28511.93
Roger Evans334455.12
Rodger Kibble412812.59
Chris Mellish52064509.43
Daniel S. Paiva6372.95
Mike Reape79713.89
Donia Scott863471.58
NEIL T IPPER9121.19