<?xml version="1.0" standalone="yes"?>
<Paper uid="W98-1211">
  <Title>IH HI I I I I i I Linguistic Theory in Statistical Language Learning</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> What role does linguistics play in statistical language learning? &amp;quot;None at all!&amp;quot; might be the answer, if we ask hard-core speech-recognition professionals.</Paragraph>
    <Paragraph position="1"> But even the most nonlinguistic language model, for example a statistic word bigram model, actually relies on key concepts integral to virtually all linguistic theories. Words, for example, and the notion that sequences of words form utterances.</Paragraph>
    <Paragraph position="2"> Statistical language learning is applied to some set of data to extract a language model of some kind.</Paragraph>
    <Paragraph position="3"> This language model can serve a purely decorative purpose, but is more often than not used to process data in some way, for example to aid speech recognition. Anyone working under the pressure of producing better results, and who employs language models to this purpose, such a researchers in the field of speech recognition, will have a high incentive of incorporating useful aspects of language into his or her language models. Now, the most useful, and thus least controversial ways of describing language will, due to their usefulness, find their way into most linguistic theories and, for the very same reason, be used in models that strive to model language successfully.</Paragraph>
    <Paragraph position="4"> So what do the linguistic theories underlying various statistical language models look like? And why? It may be useful to distinguish between those aspects of linguistic theory that are incidentally in the language model, and those that are there intentionally.</Paragraph>
    <Paragraph position="5"> We will start our tour of statistical language learning by inspecting language models with &amp;quot;very little&amp;quot; linguistic content, and then proceed to analyse increasingly more linguistic models, until we end with models that are entirely linguistic, in the sense that they are pure grammars, associated with no statistical parameters.</Paragraph>
  </Section>
class="xml-element"></Paper>