<?xml version="1.0" standalone="yes"?>
<Paper uid="W06-2922">
  <Title>Experiments with a Multilanguage Non-Projective Dependency Parser</Title>
  <Section position="7" start_page="167" end_page="168" type="evalu">
    <SectionTitle>
5 Performance
</SectionTitle>
    <Paragraph position="0"/>
    <Paragraph position="2"> Entropy, which is very fast both in learning and classification. On a 2.8 MHz Pentium Xeon PC, the learning time is about 15 minutes for Portuguese and 4 hours for Czech. Parsing is also very fast, with an average throughput of 200 sentences per second: Table 1 reports parse time for parsing each whole test set. Using Memory Based Learning increases considerably the parsing time, while as expected learning time is quite shorter. On the other hand MBL achieves an improvement up to 5% in accuracy, as shown in detail in Table 1.</Paragraph>
    <Paragraph position="3"> zou moeten worden gemaakt in zou gemaakt moeten worden in Vetsinu techto pristroju lze take pouzivat nejen jako fax , ale  For details on the CoNLL-X shared task and the measurements see (Buchholz, et al. 2006).</Paragraph>
  </Section>
  <Section position="8" start_page="168" end_page="168" type="evalu">
    <SectionTitle>
6 Experiments
</SectionTitle>
    <Paragraph position="0"> I performed several experiments to tune the parser.</Paragraph>
    <Paragraph position="1"> I also tried alternative machine learning algorithms, including SVM, Winnow, Voted Perceptron. null The use of SVM turned out quite impractical since the technique does not scale to the size of training data involved: training an SVM with such a large number of features was impossible for any of the larger corpora. For smaller ones, e.g. Portuguese, training required over 4 days but produced a bad model which could not be used (I tried both the TinySVM (Kudo 2002) and the LIBSVM (Chang and Lin 2001) implementations).</Paragraph>
    <Paragraph position="2"> Given the speed of the Maximum Entropy classifier, I explored whether increasing the number of features could improve accuracy. I experimented adding various features controlled by the parameters above: none appeared to be effective, except the addition of the previous action.</Paragraph>
    <Paragraph position="3"> The classifier returns both the action and the label to be assigned. Some experiments were carried out splitting the task among several specialized classifiers. I experimented with: 1. three classifiers: one to decide between Shift/Reduce, one to decide which Reduce action and a third one to choose the dependency in case of Left/Right action 2. two classifiers: one to decide which action to perform and a second one to choose the dependency in case of Left/Right action None of these variants produced improvements in precision. Only a small improvement in labeled attachment score was noticed using the full, non-specialized classifier to decide the action but discarding its suggestion for label and using a specialized classifier for labeling. However this was combined with a slight decrease in unlabeled attachment score, hence it was not considered worth the effort.</Paragraph>
  </Section>
class="xml-element"></Paper>