<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-0308">
  <Title>Incrementality in Deterministic Dependency Parsing</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Incrementality in parsing has been advocated for at least two different reasons. The first is mainly practical and has to do with real-time applications such as speech recognition, which require a continually updated analysis of the input received so far. The second reason is more theoretical in that it connects parsing to cognitive modeling, where there is psycholinguistic evidence suggesting that human parsing is largely incremental (Marslen-Wilson, 1973; Frazier, 1987).</Paragraph>
    <Paragraph position="1"> However, most state-of-the-art parsing methods today do not adhere to the principle of incrementality, for different reasons. Parsers that attempt to disambiguate the input completely -- full parsing -- typically first employ some kind of dynamic programming algorithm to derive a packed parse forest and then applies a probabilistic top-down model in order to select the most probable analysis (Collins, 1997; Charniak, 2000). Since the first step is essentially nondeterministic, this seems to rule out incrementality at least in a strict sense. By contrast, parsers that only partially disambiguate the input -- partial parsing -- are usually deterministic and construct the final analysis in one pass over the input (Abney, 1991; Daelemans et al., 1999). But since they normally output a sequence of unconnected phrases or chunks, they fail to satisfy the constraint of incrementality for a different reason.</Paragraph>
    <Paragraph position="2"> Deterministic dependency parsing has recently been proposed as a robust and efficient method for syntactic parsing of unrestricted natural language text (Yamada and Matsumoto, 2003; Nivre, 2003). In some ways, this approach can be seen as a compromise between traditional full and partial parsing. Essentially, it is a kind of full parsing in that the goal is to build a complete syntactic analysis for the input string, not just identify major constituents. But it resembles partial parsing in being robust, efficient and deterministic. Taken together, these properties seem to make dependency parsing suitable for incremental processing, although existing implementations normally do not satisfy this constraint. For example, Yamada and Matsumoto (2003) use a multi-pass bottom-up algorithm, combined with support vector machines, in a way that does not result in incremental processing.</Paragraph>
    <Paragraph position="3"> In this paper, we analyze the constraints on incrementality in deterministic dependency parsing and argue that strict incrementality is not achievable. We then analyze the algorithm proposed in Nivre (2003) and show that, given the previous result, this algorithm is optimal from the point of view of incrementality. Finally, we evaluate experimentally the degree of incrementality achieved with the algorithm in practical parsing.</Paragraph>
  </Section>
class="xml-element"></Paper>