<?xml version="1.0" standalone="yes"?>
<Paper uid="P06-3009">
  <Title>Sydney, July 2006. c(c)2006 Association for Computational Linguistics Integrated Morphological and Syntactic Disambiguation for Modern Hebrew</Title>
  <Section position="9" start_page="53" end_page="53" type="concl">
    <SectionTitle>
6 Conclusion
</SectionTitle>
    <Paragraph position="0"> Traditional approaches for devising parsing models, smoothing techniques and evaluation metrics are not well suited for MH, as they presuppose 13The lack of head marking, for instance, precludes the use of lexicalized models a la (Collins, 1997).</Paragraph>
    <Paragraph position="1"> 14SBJ for subject, OBJ for object, COM for complement, etc. (Sima'an et al., 2001).</Paragraph>
    <Paragraph position="2"> 15The size of our treebank is less than 30% of the Arabic Treebank, and less than 10% of the WSJ Penn Treebank.</Paragraph>
    <Paragraph position="3"> separate levels of processing. Different languages mark regularities in their surface structures in different ways English encodes regularities in word order, while MH provides useful hints about grammatical relations in its derivational and in ectional morphology. In the future we intend to develop more sophisticated models implementing closer interaction between morphology and syntax, by means of which we hope to boost parsing accuracy and improve morphological disambiguation.</Paragraph>
    <Paragraph position="4"> Acknowledgments I would like to thank Khalil Sima'an for supervising this work, Remko Scha, Rens Bod and Jelle Zuidema for helpful comments, and Alon Itai, Yoad Winter and Shuly Wintner for discussion. The Knowledge Center for Hebrew Processing provided corpora and tools, and Roy Bar-Haim provided knowledge and technical support for which I am grateful. This work is funded by the Netherlands Organization for Scienti c Research (NWO) grant 017.001.271.</Paragraph>
  </Section>
class="xml-element"></Paper>