<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-1660">
  <Title>Empirical Study on the Performance Stability of Named Entity Recognition Model across Domains</Title>
  <Section position="7" start_page="515" end_page="515" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> Efficient and robust NER model is very important in practice. This paper provides an empirical study on the impact of training data size and domain information on the performance stability of NER. Experimental results show that it is difficult to significantly enhance the performance when the training data size is above a certain threshold. The threshold of the training data size varies with domains. The performance stability of each NE type recognition also varies with domains. The large-scale corpus statistic data also show that NE types have different distribution across domains. These empirical investigations provide useful hints for enhancing the performance stability of NER models across domains with less efforts. In order to enhance the NER performance across domains, we present an informative training sample selection method. Experimental results show that the performance is significantly enhanced by using informative training samples.</Paragraph>
    <Paragraph position="1"> In the future, we'd like to focus on further exploring more effective methods to adapt NER model to a new domain with much less efforts, time and performance degrading.</Paragraph>
  </Section>
class="xml-element"></Paper>