Machine translation: a knowledge-based approach
This is the first book devoted exclusively to knowledge-based machine translation. While most approaches to the machine translation for natural languages seek ways to translate source language texts into target language texts without full understanding of the text, knowledge-based machine translation is based on extracting and representing the meaning of the source text. It is scientifically the most challenging approach to the task of machine translation, and significant progress has been achieved within it in recent years.
The authors introduce the general paradigm of knowledge-based MT, survey major recent developments, compare it with other approaches and present a paradigmatic view of its component processes
Results 1-3 of 59
Research and development efforts tend to grow significantly with the depth of
source language analysis desired in a system. Therefore, some projects perform
a relatively shallow analysis in the hope that it will actually prove sufficient for
providing posteditable translations. In other projects, one of the central tasks is to
determine the meaning of the source language text, that is, to analyze the input
texts very deeply. The adherents of shallow analysis sometimes justify their
choice not ...
It is expected that the target language texts produced from such improved
meaning representations will be of comparable quality with translations produced
by human translators knowledgeable in the subject domain. The flow of control in
KBMT systems is illustrated in figure 1.5. In brief, the input or source language
text is processed by a battery of text analysis programs. Using the knowledge
recorded in the source language grammar and lexicon, these programs (after
several stages ...
Outside the field of artistic texts — poetry and fiction — preservation of the
syntactic form of the source text in translation is truly unnecessary and, in fact,
superfluous because the meaning and use of, say, passive voice constructions in
a source and a target language should not necessarily be identical. Direct
structural correspondences between certain pairs of languages can be exploited
in MT systems of a particular type, but they should be treated as idiosyncratic
occasions rather ...
What people are saying - Write a review
Treatment of Meaning in MT Systems
The Concept of Interlingua
Lexicography and Knowledge Acquisition
6 other sections not shown
Other editions - View all
Grading Knowledge: Extracting Degree Information from Texts, Issue 1744
No preview available - 1999