File Information

File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/concl/06/n06-1030_concl.xml

Size: 972 bytes

Last Modified: 2025-10-06 13:55:09

<?xml version="1.0" standalone="yes"?>
<Paper uid="N06-1030">
  <Title>Learning Pronunciation Dictionaries Language Complexity and Word Selection Strategies</Title>
  <Section position="7" start_page="238" end_page="238" type="concl">
    <SectionTitle>
6 Conclusions
</SectionTitle>
    <Paragraph position="0"> While this work does not definitively answer the question of &amp;quot;how may words to learn the rules,&amp;quot; we have developed ways of characterizing language complexity, which can guide developers.</Paragraph>
    <Paragraph position="1"> We've devised a word selection strategy that appears to perform better than the (surprisingly high) standard set by randomly selection. Further improvements are possible by incorporating knowledge of word alignment and rule sequencing errors. By design, our strategy is biased towards short words over long, thereby being &amp;quot;nice&amp;quot; to lexicon developers - our original objective.</Paragraph>
  </Section>
class="xml-element"></Paper>
Download Original XML