<?xml version="1.0" standalone="yes"?>
<Paper uid="P06-1045">
  <Title>Selection of Effective Contextual Information for Automatic Synonym Acquisition</Title>
  <Section position="3" start_page="0" end_page="353" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Lexical knowledge is one of the most important resources in natural language applications, making it almost indispensable for higher levels of syntactical and semantic processing. Among many kinds of lexical relations, synonyms are especially useful ones, having broad range of applications such as query expansion technique in information retrieval and automatic thesaurus construction.</Paragraph>
    <Paragraph position="1"> Various methods (Hindle, 1990; Lin, 1998; Hagiwara et al., 2005) have been proposed for synonym acquisition. Most of the acquisition methods are based on distributional hypothesis (Harris, 1985), which states that semantically similar words share similar contexts, and it has been experimentally shown considerably plausible.</Paragraph>
    <Paragraph position="2"> However, whereas many methods which adopt the hypothesis are based on contextual clues concerning words, and there has been much consideration on the language models such as Latent Semantic Indexing (Deerwester et al., 1990) and Probabilistic LSI (Hofmann, 1999) and synonym acquisition method, almost no attention has been paid to what kind of categories of contextual information, or their combinations, are useful for word featuring in terms of synonym acquisition.</Paragraph>
    <Paragraph position="3"> For example, Hindle (1990) used co-occurrences between verbs and their subjects and objects, and proposed a similarity metric based on mutual information, but no exploration concerning the effectiveness of other kinds of word relationship is provided, although it is extendable to any kinds of contextual information.</Paragraph>
    <Paragraph position="4"> Lin (1998) also proposed an information theory-based similarity metric, using a broad-coverage parser and extracting wider range of grammatical relationship including modifications, but he didn't further investigate what kind of relationships actually had important contributions to acquisition, either. The selection of useful contextual information is considered to have a critical impact on the performance of synonym acquisition. This is an independent problem from the choice of language model or acquisition method, and should therefore be examined by itself.</Paragraph>
    <Paragraph position="5"> The purpose of this study is to experimentally investigate the impact of contextual information selection for automatic synonym acquisition. Because nouns are the main target of  synonym acquisition, here we limit the target of acquisition to nouns, and firstly extract the co-occurrences between nouns and three categories of contextual information -- dependency, sentence co-occurrence, and proximity -- from each of three different corpora, and the performance of individual categories and their combinations are evaluated. Since dependency and modification relations are considered to have greater contributions in contextual information and in the dependency category, respectively, these categories are then broken down into smaller categories to examine the individual significance.</Paragraph>
    <Paragraph position="6"> Because the consideration on the language model and acquisition methods is not the scope of the current study, widely used vector space model (VSM), tf*idf weighting scheme, and cosine measure are adopted for similarity calculation. The result is evaluated using two automatic evaluation methods we proposed and implemented: discrimination rate and correlation coefficient based on the existing thesaurus WordNet.</Paragraph>
    <Paragraph position="7"> This paper is organized as follows: in Section 2, three kinds of contextual information we use are described, and the following Section 3 explains the synonym acquisition method. In Section 4 the evaluation method we employed is detailed, which consists of the calculation methods of reference similarity, discrimination rate, and correlation coefficient. Section 5 provides the experimental conditions and results of contextual information selection, followed by dependency and modification selection. Section 6 concludes this paper.</Paragraph>
  </Section>
class="xml-element"></Paper>