<?xml version="1.0" standalone="yes"?>
<Paper uid="A97-1029">
  <Title>Nymble: a High-Performance Learning Name-finder</Title>
  <Section position="3" start_page="0" end_page="194" type="intro">
    <SectionTitle>
2. Background
</SectionTitle>
    <Paragraph position="0"/>
    <Section position="1" start_page="0" end_page="194" type="sub_section">
      <SectionTitle>
2.1 Name-finding as an Information-
theoretic Problem
</SectionTitle>
      <Paragraph position="0"> The basic premise of the approach is to consider the raw text encountered when decoding as though it had passed through a noisy channel, where it had been originally marked with named entities. ~ The job of the generative model is to model the original process which generated the name-class-annotated words, before they went through the noisy channel.</Paragraph>
      <Paragraph position="1"> More formally, we must find the most likely sequence of name-classes (NC) given a sequence of words (W):</Paragraph>
      <Paragraph position="3"> In order to treat this as a generative model (where it generates the original, name-class-annotated words),</Paragraph>
      <Paragraph position="5"> and since the a priori probability of the word sequence--the denominator--is constant for any given sentence, we can maxi-mize Equation 2,2 by maximizing the numerator alone.</Paragraph>
    </Section>
    <Section position="2" start_page="194" end_page="194" type="sub_section">
      <SectionTitle>
2.2 Previous
</SectionTitle>
      <Paragraph position="0"> Approaches to Name-finding null Previous approaches have typically used manually constructed finite state patterns (Weischedel, 1995, Appelt et al., 1995). For START-OF-SENTENCE. every new language and every ~ new class of new information to spot, one has to write a new set of rules to cover the new language and to cover the new class of information. A finite-state pattern rule attempts to match against a sequence of tokens (words), in much the same way as a general regular expression matcher.</Paragraph>
      <Paragraph position="1"> In addition to these finite-state pattern approaches, a variant of Brill rules has been applied to the problem, as outlined in (Aberdeen et al., 1995).</Paragraph>
    </Section>
    <Section position="3" start_page="194" end_page="194" type="sub_section">
      <SectionTitle>
2.3 Interest in Problem and Potential
Applications
</SectionTitle>
      <Paragraph position="0"> The atomic elements of information extraction-indeed, of language as a whole---could be considered the who, where, when and how much in a sentence.</Paragraph>
      <Paragraph position="1"> A name-finder performs what is known as surface- or lightweight-parsing, delimiting sequences of tokens that answer these important questions. It can be used as the first step in a chain of processors: a next level of processing could relate two or more named entities, or perhaps even give semantics to that relationship using a verb. In this way, further processing could discover the &amp;quot;what&amp;quot; and &amp;quot;how&amp;quot; of a sentence or body of text.</Paragraph>
      <Paragraph position="2"> Furthermore, name-finding can be useful in its own right: an Internet query system might use name-finding to construct more appropriately-formed queries: &amp;quot;When was Bill Gates born?&amp;quot; could yield the query &amp;quot;BS.1\]_ Gat:es&amp;quot;+born. Also, name-finding can be directly employed for link analysis and other information retrieval problems.</Paragraph>
    </Section>
  </Section>
class="xml-element"></Paper>