<?xml version="1.0" standalone="yes"?>
<Paper uid="W00-1412">
  <Title>Incr en ntal, Eventoneeptua fization.:and Natural Language Generation in Monitoring Environments</Title>
  <Section position="2" start_page="0" end_page="85" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Systems that generate natural language descriptions of what happens in a dynamically changing world can be improved substantially by working incrementally. Incrementality enhances the overall quality of the systems for three reasons: (1) The dynamic nature of a continuous stream of input information can be handled more directly and, therefore, easier. (2) Incremental systems are capable of producing fluent speech, i.e. speech without artificial auditory gaps. (3) Parallelism that comes with incrementality makes better use of the available resources.</Paragraph>
    <Paragraph position="1"> Furthermore, Reiter (1994), who reviews the * architecture of some models of natural language generation, shows that psycholinguistic and engineering approaches often result in systems, which are similar in crucial respects. In this paper we ground on two of these common aspects, namely the distinction between what-to-say and how-tosay (De Smedt, Horacek &amp; Zock, 1996) and the use of a pipeline architecture, which divides the generation process &amp;quot;into multiple modules, with information flowing in a 'pipeline' fashion from one module to the next&amp;quot; (Reiter, 1994). Reiter states that these architectures do not require modules to work in parallel; if parallelism is used one has an incremental model, cf. De Smedt &amp; Kernpen (I 987), Ferreira (1996).</Paragraph>
    <Paragraph position="2"> The primary research topic of the ConcEv ~ project is the what-to-say component, in which the content of utterances is planned (Reiter's content planning, in contrast to the language specific sentence planning component). We use the terminology of Levelt (1989), who calls the first component the conceptualizer, the second the formulator. These modules interact via preverbal messages, which are propositional, non-verbal representations of the utterance built up by the conceptualizer. They are transformed by the formulator into linguistic structures for spoken or written output. Besides considering high level communicative goals (macroplanning), which are in the focus of most computational approaches to the what-to-say component, e.g. De Smedt &amp; Kempen (1987), McKeown (1985), Hovy (1993), Chu-Carroll &amp; Carberry (1998), Radev &amp; McKeown (1998), the type of information to be verbalized also determines the processes of conceptualization on the level of microplanning, cf. Levelt (1989). Thus, the traditional top-down approaches have to be combined with bottom-up data-driven approaches of text planning (Marcu, 1997~. The conceptualizer that is described in detail in section 3 fits the pipeline architecture on a coarse level, but integrates on finer levels the ideas of functional modules (Cahill et al., 1999).</Paragraph>
    <Paragraph position="3"> In the present paper we focus on the task of  generating verbal descriptions of continuously in-.</Paragraph>
    <Paragraph position="4"> coming input from a changing physical world (see section 2, for similar settings cf. Neumann &amp; Novak (1983) and Andr6, Herzog &amp; Rist (1988)).</Paragraph>
    <Paragraph position="5"> This specific task requires an incremental pipeline architecture--as there are certain steps that have to be carried out in a specific order--and, additionally, these steps can be organized in such a .... insights .into _psy.cho.lJnguistic.aspects _of natural language processing. Our implementation thus simulates aspects of behavior, e.g. the effects differing time pressure has on verbalizations.</Paragraph>
  </Section>
class="xml-element"></Paper>