<?xml version="1.0" standalone="yes"?>
<Paper uid="C02-1088">
  <Title>Unsupervised Named Entity Classification Models and their Ensembles</Title>
  <Section position="6" start_page="3" end_page="3" type="concl">
    <SectionTitle>
4 Conclusion
</SectionTitle>
    <Paragraph position="0"> We proposed an unsupervised learning model for classifying the named entities. This model used a training set, built automatically by a small-scale NE dictionary and an unlabeled corpus, instead of a hand-tagged training set for learning. The experimental result showed 73.16% in precision and 72.98% in recall for Korean news articles. This means that it is possible to classify named entities without the cost for building a large hand-tagged training corpus or a lot of rules.</Paragraph>
    <Paragraph position="1"> The learning for classification was progressed by the ensemble of three different learning methods. Then the ensemble of various learning methods brings a better result than each individual learning method.</Paragraph>
  </Section>
class="xml-element"></Paper>