<?xml version="1.0" standalone="yes"?>
<Paper uid="P06-1145">
  <Title>Sydney, July 2006. c(c)2006 Association for Computational Linguistics Time Period Identification of Events in Text Taichi Noro + Takashi Inui ++ Hiroya Takamura ++</Title>
  <Section position="8" start_page="1159" end_page="1159" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> In our study, we proposed a method for identifying when an event in text occurs. We succeeded in using a semi-supervised method, the Naive Bayes Classifier enhanced by the EM algorithm, with a small amount of labeled data and a large amount of unlabeled data. In order to avoid the class imbalance problem, we used a 2-step classifier, which first filters out time-unknown sentences and then classifies the remaining sentences into one of 4 classes. The proposed method outperformed the simple 1-step method.</Paragraph>
    <Paragraph position="1"> We obtained 86.4% of accuracy that exceeds the existing method and the baseline method.</Paragraph>
  </Section>
class="xml-element"></Paper>