<?xml version="1.0" standalone="yes"?>
<Paper uid="W99-0109">
  <Title>O Q O O O O O @ @ O O O O O O O O O O @ O O O O O O O O O @ O O</Title>
  <Section position="8" start_page="78" end_page="79" type="concl">
    <SectionTitle>
4 Conclusions and future work
</SectionTitle>
    <Paragraph position="0"> CT has developed primarily as a tool for analysing the structure of a given text and identifying the most likely candidates for anaphora resolution. In this paper we sought to deterdeg mine whether the principles underlying the constraints and rules of the theory can be Uturned round&amp;quot; and used as planning operators for generating coherent text,. As a side-effect of this enterprise we have articulated a ~streamlined&amp;quot; formulation of CT in terms of the principles of salience and cohesion, and argued that the preferences for the different transition types emerge in a partial ordering f~om the interaction between these principles. These principles are rather heterogenous, a fact which is obscured by combining them in the transition definitions, and can be implemented as encapsulated tasks distributed between text planning,  sentence planning and RE generation. It may turn out that individual components such as the cohesion principle do not need to be explicitly stated but emerge as by-products from higher-level text planning.</Paragraph>
    <Paragraph position="1"> As noted at various points in this paper CT has never been more than partially implemented in NLG systems. This may be due to a belief on the part of NLG practitioners that CT gets things the wrong way round, by relying on surface grammatical realisation to determine the * centre of attention in an utterance. If this belld is commonly held (and anecdotal evidence suggests that it is), I argue that it is mistaken. In interpretation systems the principles of CT guide the system in identifying the centre of attention and in choosing likely referents for anaphors. In NLG systems, if there is a notion of &amp;quot;topic&amp;quot; or &amp;quot;theme = this should be designated by the text planner, while the CT rules allow the sentence planner to promote this entity to salience to keep it as the user's &amp;quot;centre of attention&amp;quot;. null However, a more fuDd~mental explanation for the neglect of CT in the generation literature is provided by the fact that a faithful implementation in a pipelined system turns out to require an independent way of designating the central entity in a proposition, and this itself is a problem which has not had much attention in the development of NLG systems 2. So the next stage in this research will Concentrate on developing a characterisation of Cb based on semantic content and information structure, taking account of e.g. the proposals of Strube and Hahn (1996) and experimenting with optlmisation algorithms as discussed in section 3.1.2 above.</Paragraph>
    <Paragraph position="2"> As mentioned above the Reiter model has been questioned by members of the RAGS project in Brighton and Edinburgh who are actively engaged in developing a &amp;quot;Teference&amp;quot; archi tecture for NLG. (See Cahlll et al., 1999, RAGS 1999.) To date' the group has concentrated on specifying the data structures which are required at various stages of the generation task and has identified a number of discrete functions such as rhetorical structuring, aggregation, ~Thk appHm at lea~ to applied systems;, see CahiU * and Reape 1998. One exception appears to be the GOS-. SIP system described in Caragno and Iordaaskaja 1989. coherence etc., without specifying a strict order for the execution of these functions. It is too early to assess how the proposals of this paper would fit into the RAGS scheme, but I anticipate that, as with the Reiter architecture, the conclusion would be that referential coherence is not the task of a discrete module but imposes constraints on a number of different modules.</Paragraph>
    <Paragraph position="3"> It has been noted that the way the &amp;quot;realise&amp;quot; relation is interpreted can have significant implications for the coherence of a text as measured by CT, and that corpus analysis has often concentrated ondirectly realised Cb's. An exception is Hahn, Strube and Markert's (1996) treatment of bridging reference, or &amp;quot;textual ellipsis&amp;quot; in their terminology. As these authors note there have been rather few implemented systems which are able to interpret bridging references in a principled way, and research in NLG is particularly weak in this area. SO, a faithful implementation of CT in generation systems will depend in part on progress in the generation of bridging references. This is an area for future research.</Paragraph>
    <Paragraph position="4"> .... It is intended that the procedures described in &amp;quot;this&amp;quot; paper will be implemented in ICONOCLAST, an authoring tool which enables domain experts to create a knowledge base through a sequence of interactive choice-- and generates hiexarchitally structured text according to various stylistic constraints (See Power and Scott 1998).</Paragraph>
  </Section>
class="xml-element"></Paper>