<?xml version="1.0" standalone="yes"?>
<Paper uid="N03-1002">
  <Title>Japanese Named Entity Extraction with Redundant Morphological Analysis</Title>
  <Section position="8" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5 Conclusions
</SectionTitle>
    <Paragraph position="0"> The proposed NE extraction method achieves F-measure 87.21 on CRL NE data. This is the best result in the previously reported systems. We made use of character level information with redundant outputs of a statistical morphological analyzer in an SVM-based chunker. It copes with the word unit problem in NE extraction. Furthermore, the method is robust for both errors of the morphological analyzer and occurences of unknown words, because character level prefixes and suffixes of NEs are clues for finding them. Fragments of possible words are used as features by the redundant morphological analysis. Though we tested this method only with Japanese, the method is applicable to any other languages that have word unit problem in NE extraction.</Paragraph>
  </Section>
class="xml-element"></Paper>