<?xml version="1.0" standalone="yes"?>
<Paper uid="W97-0208">
  <Title>Sense Tagging: Semantic Tagging with a Lexicon</Title>
  <Section position="3" start_page="47" end_page="47" type="intro">
    <SectionTitle>
2 Recent Word Sense
</SectionTitle>
    <Paragraph position="0"> Disambiguation algorithms Recent word sense disambignation (WSD) algorithms can be categorised into two broad types:  1. WSD using information in an explicit lexicon. This is usually a Machine Readable Dictionary (MRD) such as the Longman Dictionary o\] Contemporary English (LDOCE) (Procter, 1978), WordNet (Miller (Ed.), 1990) or handcrafted. Recent examples of this work include (Bruce and Guthrie, 1992), (Bruce and Wiebe, 1994), (McRoy, 1992).</Paragraph>
    <Paragraph position="1"> 2. WSD using information gained from training on some corpus. This approach can be further subm null m classified: (a) Supervised training, where information is gathered from corpora which have already been semantically disambiguated. As such corpora are hard to obtain, usually requiring expensive hand-tagging, research in this area has concentrated on other forms of lexical ambiguities, eg. (Gale, Church, and Yarowsky, 1992).</Paragraph>
    <Paragraph position="2"> (b) Unsupervised training, where information  is gathered from raw corpora which has not been semantically disambiguated. The best examples of this approach has been the resent work of Yarowsky - (Yarowsky, 1992), (Yarowsky, 1993), (Yarowsky, 1995). These approaches are not mutually exclusive and there are, of course, some hybrid cases, for example Luk (Luk, 1995) uses information in MRD definitions (approach 1) and statistical information from untagged corpora (approach 2b).</Paragraph>
  </Section>
class="xml-element"></Paper>