<?xml version="1.0" standalone="yes"?>
<Paper uid="W02-2007">
  <Title>Language Independent NER using a Unified Model of Internal and Contextual Evidence</Title>
  <Section position="10" start_page="0" end_page="0" type="concl">
    <SectionTitle>
10. Results
</SectionTitle>
    <Paragraph position="0"> We compare the results of two variants of the described model on the development and test sets provided (Table 1). The first one uses only exemplar entity and context seeds extracted from the training corpus. The second also employs POS information to rule out unlikely entity candidates.</Paragraph>
    <Paragraph position="1"> The system was built and tested initially utilizing only the provided Spanish data. The parameters were estimated using an 80/20 split of the training data (esp.train and ned.train). The dev-test data (testa) were not used during the parameter estimation phase. The programs were run once on the final test data (files testb). We allocated only one person-day to adapt the system for Dutch and tune the parameters to this language in order to show functional language independence. We opted not to make a detailed study of parameter variation on test data to avoid any potential for tuning to this resource and preserve its value for future system development.</Paragraph>
    <Paragraph position="2"> The following table further details the types of errors made by the algorithm (full system on Spanish dev-set). a128 a100 represents the number of over-generated and under-generated entities in the precision and recall rows (respectively). a128a111a129 represents the number of entities with correctly identified boundaries, but wrong classifications.</Paragraph>
    <Paragraph position="3">  Because our system takes seed lists rather than annotated text as input, additional entity lists can be used by the system. By employing such lists of countries, major cities, frequent person names and major companies (extracted from the web), significant improvements can be obtained (preliminary tests show as much as 2.5 F-measure improvement on a 80/20 split of the training data in Dutch).</Paragraph>
    <Paragraph position="4"> 11. Conclusion This paper has presented and evaluated an extended bootstrapping model based on Cucerzan and Yarowsky (1999) that uses a unified framework of both entity internal and contextual evidence. Start-Spanish without POS information with POS information  ned.testa) and on the test sets (files esp.testb and ned.testb) ing only with entity and context seeds extracted from training data and the addition of part-of-speech information, system performance exceeds 77 and 72 F-measure for Spanish and Dutch respectively.</Paragraph>
  </Section>
class="xml-element"></Paper>