<?xml version="1.0" standalone="yes"?>
<Paper uid="W00-0723">
  <Title>Learning IE Rules for a Set of Related Concepts</Title>
  <Section position="2" start_page="117" end_page="117" type="ackno">
    <SectionTitle>
4 Evaluation
</SectionTitle>
    <Paragraph position="0"> EVIUS has been tested on the mycological domain. A set of 68 Spanish mycological documents (covering 9800 words corresponding to 1360 lemmas) has been used. 13 of them have been kept for testing and the others for training. The target ontology consisted of 14 concepts and 24 relations.</Paragraph>
    <Paragraph position="1"> Several experiments have been carried out with different training sets. Results of the initial rule set for the colour concept 1deg are presented in table 1.</Paragraph>
    <Paragraph position="2"> Out of 34 in the 350 initial rule set, one of the most relevant learned rules is11: Col our ( A, B ) :-has_h ypern ym_OOO17586n ( B ) , has_hypernym_O3464624n (A), brother (A, B).</Paragraph>
    <Paragraph position="3"> Table 2 shows the results of adding pseudo-examples to the 35012 training set and using the algorithm in section 3. This was tested with a = 0.01 (two iterations are enough, 351 and 352) and 5 pseudo-examples for each uncovered case. The algorithm returns the rule set produced in the first iteration due to the fact that ~F1T13&gt; 0.01 between the first and the second iterations. Higher results can be generated when using lower values for a.</Paragraph>
    <Paragraph position="4"> Although no direct comparison with other systems is possible due to the domain and language used, our results can be considered state- null to the initial training set with 35 documents.</Paragraph>
    <Paragraph position="5"> of-the-art regarding similar MUC competition tasks.</Paragraph>
  </Section>
class="xml-element"></Paper>