<?xml version="1.0" standalone="yes"?>
<Paper uid="P06-2039">
  <Title>Parsing Aligned Parallel Corpus by Projecting Syntactic Relations from Annotated Source Corpus</Title>
  <Section position="8" start_page="307" end_page="307" type="concl">
    <SectionTitle>
7 Concluding Remarks
</SectionTitle>
    <Paragraph position="0"> The present work focuses on development of Example based parsing scheme for a pair of languages in general, and for English to Hindi in particular. null Although the current work is motivated by (Hwa et al., 2005), the algorithm proposed herein provides a more generalized version of the projection algorithm by making use of some target language specific rules while projecting links. This provide more flexibility in the projection algorithm. The flexibility comes from the fact that unlike DPA the algorithm can project links from the source language to the target language even if the translations are not literal. Use of rules at the projection level gives more robust parsing and reduces the need of post-editing. The proposed scheme should work for other target languages also provided the relevant rules can be identified. Further, since LG can be converted to Dependency Grammar (DG) (Sleator and Temperley, 1991), this work can be easily extended for languages for which DG implementation is available.</Paragraph>
    <Paragraph position="1"> At present, we have focused on developing parsing scheme for simple sentences. Work has to be done to parse complex sentences. Once a sizeable parsed corpus is generated, it can be used for developing the parser for a target language using bootstrapping. We are currently working on these lines for developing a Hindi parser.</Paragraph>
  </Section>
class="xml-element"></Paper>