<?xml version="1.0" standalone="yes"?>
<Paper uid="P06-1099">
  <Title>You Can't Beat Frequency (Unless You Use Linguistic Knowledge) - A Qualitative Evaluation of Association Measures for Collocation and Term Extraction</Title>
  <Section position="3" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Research on domain-specific automatic term recognition (ATR) and on general-language collocation extraction (CE) has gone mostly separate ways in the last decade although their underlying procedures and goals turn out to be rather similar. In both cases, linguistic filters (POS taggers, phrase chunkers, (shallow) parsers) initially collect candidates from large text corpora and then frequency- or statistics-based evidence or association measures yield scores indicating to what degree a candidate qualifies as a term or a collocation. While term mining and collocation mining, as a whole, involve almost the same analytical processing steps, such as orthographic and morphological normalization, normalization of term or collocation variation etc., it is exactly the measure which grades termhood or collocativity of a candidate on which alternative approaches diverge.</Paragraph>
    <Paragraph position="1"> Still, the output of such mining algorithms look similar. It is typically constituted by a ranked list on which, ideally, the true terms or collocations are placed in the top portion of the list, while the non-terms / non-collocations occur in its bottom portion.</Paragraph>
    <Paragraph position="2"> While there have been lots of approaches to come up with a fully adequate ATR/CE metric (cf. Section 2), we have made observations in our experiments that seem to indicate that simplicity rules, i.e., frequency of occurrence is the dominating factor for the ranking in the result lists even when much smarter statistical machinery is employed. In this paper, we will discuss data which reveals that purely statistics-based measures exhibit virtually no difference compared with frequency of occurrence counts, while linguistically more informed measures do reveal such a marked difference - for the problem of term and collocation mining at least.</Paragraph>
  </Section>
class="xml-element"></Paper>