<?xml version="1.0" standalone="yes"?>
<Paper uid="P01-1041">
  <Title>Japanese Named Entity Recognition based on a Simple Rule Generator and Decision Tree Learning</Title>
  <Section position="10" start_page="91" end_page="91" type="concl">
    <SectionTitle>
5 Conclusions
</SectionTitle>
    <Paragraph position="0"> As far as we can tell, Japanese NE recognition technology has not yet matured. Conventional decision tree systems have not shown good performance. The maximum entropy method is competitive, but adding more training data causes problems. In this paper, we presented an alternative method based on decision tree learning and longest match. According to our experiments, this method's performance is comparable to that of the maximum entropy system, and it can be trained more efficiently. We hope our method can be applicable to other languages.</Paragraph>
  </Section>
class="xml-element"></Paper>