<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-0824">
  <Title>Multi-Component Word Sense Disambiguationa0</Title>
  <Section position="6" start_page="0" end_page="0" type="concl">
    <SectionTitle>
5 Results
</SectionTitle>
    <Paragraph position="0"> Table 1 presents results from a set of experiments performed by cross-validation on the training data, for several nouns and verbs.For 37 nouns and verbs, out of 52, the two-component model was more accurate than the flat model5. We used the results from these experiments to set, separately for each word, the parameters a68 , which was equal to 13.9 on average, and a75 a70 . For adjectives we only set the parameter a68 and used the standard &amp;quot;flat&amp;quot; perceptron. For each word in the task we separately trained one classifier. The system accuracy on the unseen test set is summarized in the following table:  Overall the system has the following advantages over that of (Ciaramita et al., 2003). Selecting the external training data based on the most similar a60 synsets has the advantage, over using supersenses, of generating an equivalent amount of additional data for each word sense. The additional data for each synset is also more homogeneous, thus the  with different values for a3 a4 , the multicomponent model would achieve even better performances.</Paragraph>
    <Paragraph position="1"> model should have less variance6. The multicomponent architecture is simpler and has an obvious convergence proof. Convergence is faster and training is efficient. It takes less than one hour to build and train all final systems and generate the complete test results. We used the averaged version of the perceptron and introduced an adjustable parameter a75 to weigh each component's contribution separately.</Paragraph>
  </Section>
class="xml-element"></Paper>