<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-0506">
  <Title>Sydney, July 2006. c(c)2006 Association for Computational Linguistics Taxonomy Learning using Term Specificity and Similarity</Title>
  <Section position="4" start_page="41" end_page="41" type="intro">
    <SectionTitle>
D
</SectionTitle>
    <Paragraph position="0"> The strength of this method lies in its ability to adopt different optimal features for term specificity and term similarity. Most of current researches relied on single feature such as adjectives of terms, verb-argument relation, or co-occurrence ratio in documents according to their methods. Firstly, we analyze characteristics of features for taxonomy learning in view of term specificity and term similarity to show that the features embed characteristics of specificity and similarity, and finally apply optimal features to our method.</Paragraph>
    <Paragraph position="1"> Additionally we tested inside information of terms to measure term specificity and similarity.</Paragraph>
    <Paragraph position="2"> As multiword terms cover the larger part of technical terms, lexical components are featuring information representing semantics of terms (Cerbah, 2000).</Paragraph>
    <Paragraph position="3"> The remainder of this paper is organized follows. Characteristics of term specificity are described in Section 2, while term similarity and its features are addressed in Section 3. Our taxonomy learning method is discussed in Section 4. Experiment and evaluation are discussed in Section 5, and finally, conclusions are drawn in Section 6.</Paragraph>
  </Section>
class="xml-element"></Paper>