File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/intro/06/w06-2410_intro.xml
Size: 1,326 bytes
Last Modified: 2025-10-06 14:04:07
<?xml version="1.0" standalone="yes"?> <Paper uid="W06-2410"> <Title>Multiword Units in an MT Lexicon</Title> <Section position="2" start_page="0" end_page="0" type="intro"> <SectionTitle> 1 Introduction </SectionTitle> <Paragraph position="0"> The robustness of MT systems crucially depend on the size and quality of their lexical componenets. It is commonly recognized that word-to-word equivalents are fraught with ambiguities.</Paragraph> <Paragraph position="1"> MW units on the other hand carry, as it were, the disambiguating context with them. Hence, the more MW units in the lexicon and the longer they are, the less noisy and more robust the MT lexicon is likely to be. However, not all kinds of MW units are amenable to inclusion by itemized listing in the lexicon. The paper will focus on MW units whose structure contains slots that can be filled by more or less open ended lexical units. They are treated in paper dictionaries with the usual method of exemplification and implication, which, even if the intended extension of the set of expression is clear, is obviously not a viable option in a machine system that cannot rely on the linguistic competence and world knowledge that human readers of dictionaries are expected to bring to the job of interpreting lexical entries.</Paragraph> </Section> class="xml-element"></Paper>