<?xml version="1.0" standalone="yes"?>
<Paper uid="P99-1043">
  <Title>Mixed Language Query Disambiguation</Title>
  <Section position="5" start_page="336" end_page="337" type="evalu">
    <SectionTitle>
4 Evaluation results
</SectionTitle>
    <Paragraph position="0"> One advantage of using the artificially generated mixed-language test set is that it becomes very easy to evaluate the performance of the disambiguation/translation algorithm. We just need to compare the translation output with the original ATIS sentences.</Paragraph>
    <Paragraph position="1"> The experimental results are shown in Figure 4. The horizontal axis represents the percentage of English words in the testing data and the vertical axis represents the translation accuracy. Translation accuracy is the ratio of the number of secondary language (Chinese) words disambiguated correctly over the number of all  secondary language (Chinese) words present in the testing sentences. The three different curves represent the accuracies obtained from the base-line feature, the voting model, and the 1-best model.</Paragraph>
    <Paragraph position="2">  We can see that both voting contextual words and the 1-best contextual words are more powerful discriminant than the baseline neighboring word. The 1-best feature is most effective for disambiguating secondary language words in a mixed-language sentence.</Paragraph>
  </Section>
class="xml-element"></Paper>