File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/concl/00/c00-2145_concl.xml
Size: 2,034 bytes
Last Modified: 2025-10-06 13:52:45
<?xml version="1.0" standalone="yes"?> <Paper uid="C00-2145"> <Title>A Model of Competence for Corpus-Based Machine Translation</Title> <Section position="6" start_page="1000" end_page="1000" type="concl"> <SectionTitle> 6 Conclusion </SectionTitle> <Paragraph position="0"> Machine Tr'anslation (MT) is a lneaning preserving raapping from a source language text into a target language text. In order to enable a computer system to perform such a mapping, it is provided with a formalized theory of meaning.</Paragraph> <Paragraph position="1"> Theories of meaning are characterized by three dichotomies: they call be holistic or molecular, austere or rich and they can be fine-grained or coarscMuiucd. null A number of CBMT systems - translation memories, example-based and statistical-based nlachine translation systenls - arc examined with respect to these dichotomies. Ill a system that uses a rich theory of meaning, complex representations arc computed including morphological, syntactical and sernantical representations, while with an austere theory the system relics on the mere graphcnfic surface form of the text. In a holistie implementation meaning descriptions are derived from reference translations while in a molecular approach the meaning dcscriptions are obtained from a finite set of predefined features. In a fine-grained theory, the minimal length of a translation unit is equivalent to a mor1)heme while in a coarse-grained theory this amounts to a morphenle cluster, a phrase or a sentence.</Paragraph> <Paragraph position="2"> According to the implemented theory of meaning, one can expect to obtain high quality translations or a good covera.qc of the CBMT system.</Paragraph> <Paragraph position="3"> The more the system makes use of coarse-grained translation units, the higher is tlle expected translation quality. The more the theory uses rich representations thc more the system may achieve broad coverage. CBMT systems can be tuned to achieve cither of the two goals.</Paragraph> </Section> class="xml-element"></Paper>