<?xml version="1.0" standalone="yes"?>
<Paper uid="A00-1017">
  <Title>A Representation for Complex and Evolving Data Dependencies in Generation</Title>
  <Section position="3" start_page="0" end_page="119" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> One of the distinctive properties of natural language generation when compared with other language engineering applications is that it has to take seriously the full range of linguistic representation, from concepts to morphology, or even phonetics. Any processing system is only as sophisticated as its input allows, so while a natural language understanding system might be judged primarily by its syntactic prowess, even if its attention to semantics, pragmatics and underlying conceptual analysis is minimal, a generation system is only as good as its deepest linguistic representations. Moreover, any attempt to abstract away from individual generation systems to a more generic architectural specification faces an even greater challenge: not only are complex linguistic representations required, able to support the dynamic evolutionary development of data during the gener* Now at the MITRE Corporation, Bedford, MA, USA, cdoran@mitre, org.</Paragraph>
    <Paragraph position="1"> ation process, but they must do so in a generic and flexible fashion.</Paragraph>
    <Paragraph position="2"> This paper describes a representation developed to meet these requirements. It offers a formally well-defined declarative representation language, which provides a framework for expressing the complex and dynamic data requirements of NLG systems. The approach supports different levels of representation, mixed representations that cut across levels, partial and shared structures and 'canned' representations, as well as dynamic relationships between data at different stages in processing. We are using the approach to develop a high level data model for NLG systems as part of a generic generation architecture called RAGS 1.</Paragraph>
    <Paragraph position="3"> The framework has been implemented in the form of a database server for modular generation systems. As proof of concept of the framework, we have reimplemented an existing NLG system. The system we chose was the Caption Generation System (CGS) (Mittal et al., 1995; Mittal et al., 1998). The reimplementation involved defining the interfaces to the modules of CGS in terms of the RAGS representations and then implementing modules that had the requisite input and output representations.</Paragraph>
    <Paragraph position="4"> Generation systems, especially end-to-end, applied generation systems, have, unsurprisingly, many things in common. Reiter (1994) proposed an analysis of such systems in terms of a simple three stage pipeline. More recently, the RAGS project attempted to repeat the anal- null RAGS: Reference Architecture for Generation Systems. We would also like to acknowledge the contribution of Jo Calder to the ideas and formalisation described in this paper. In particular, parts of this paper are based on (Calder et al., 1999).</Paragraph>
    <Paragraph position="5">  ysis (Cahill et al., 1999a), but found that while most systems did implement a pipeline, they did not implement the same pipeline - different functionalities occurred in different places and different orders in different systems. In order to accommodate this result, we sought to develop an architecture that is more general than a simple pipeline, and thus supports the range of pipelines observed, as well as other more complex control regimes (see (Cahill et al., 1999a; Cahill et al., 1999b)). In this paper, we argue that supporting such an architecture requires careful consideration of the way data representations interact and develop. Any formal framework for expressing the architecture must take account of this.</Paragraph>
  </Section>
class="xml-element"></Paper>