File Information

File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/relat/06/p06-3009_relat.xml

Size: 2,185 bytes

Last Modified: 2025-10-06 14:15:59

<?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="5" start_page="50" end_page="50" type="relat">
    <SectionTitle>
3 Related Work
</SectionTitle>
    <Paragraph position="0"> As of yet there is no statistical parser for MH.</Paragraph>
    <Paragraph position="1"> Parsing models have been developed for different languages and state-of-the-art results have been reported for, e.g., English (Collins, 1997; Charniak, 2000). However, these models show impoverished morphological treatment, and they have not yet been successfully applied for MH parsing.</Paragraph>
    <Paragraph position="2"> (Sima'an et al., 2001) present an attempt to parse MH sentences based on a small, annotated corpus by applying a general-purpose Tree-gram model.</Paragraph>
    <Paragraph position="3"> However, their work presupposes correct morphological disambiguation prior to parsing.5 In order to treat morphological phenomena a few stand-alone morphological analyzers have been developed for MH.6 Most analyzers consider words in isolation, and thus propose multiple analyses for each word. Analyzers which also attempt disambiguation require contextual information from surrounding word-forms or a shallow parser (e.g., (Adler and Gabai, 2005)).</Paragraph>
    <Paragraph position="4">  A related research agenda is the development of part-of-speech taggers for MH and other Semitic languages. Such taggers need to address the segmentation of words into morphemes to which distinct morphosyntactic categories can be assigned (cf. gure 2). It was illustrated for both MH (Bar-Haim, 2005) and Arabic (Habash and Rambow, 2005) that an integrated approach towards making morphological (segmentation) and syntactic (POS tagging) decisions within the same architecture yields excellent results. The present work follows up on insights gathered from such studies, suggesting that an integrated framework is an adequate solution for the apparent circularity in morphological and syntactic processing of MH.</Paragraph>
  </Section>
class="xml-element"></Paper>
Download Original XML