<?xml version="1.0" standalone="yes"?>
<Paper uid="W05-0815">
  <Title>Experiments Using MAR for Aligning Corpora[?]</Title>
  <Section position="7" start_page="97" end_page="97" type="evalu">
    <SectionTitle>
6 Results and discussion
</SectionTitle>
    <Paragraph position="0"> The results for the alignment can be seen in Table 2. As mentioned above, there is a certain preference for recall over precision. For comparison, using GIZA++ on the split corpus yields a precision of 0.6834 and a recall of 0.5601 for a total AER of 0.3844.</Paragraph>
    <Paragraph position="1"> Note that although the definition of the task allowed to mark the alignment as either probable or sure, we marked all the alignments as sure, so precision and recall measures are given only for sure alignments.</Paragraph>
    <Paragraph position="2"> There are aspects that deserve further experimentation. The first is the split of the original corpus. It would be important to evaluate its influence, and to try to find methods of using MAR without any split at all. A second aspect of great importance is the method used for &amp;quot;flattening&amp;quot;. The way leaves of the tree are treated probably could be improved if the dictionary probabilities were somehow taken into account.</Paragraph>
  </Section>
class="xml-element"></Paper>