Transcript Translation(5) Lecture[1-1]

TRANSLATION(5)
LECTURE[1-1]
MACHINE TRANSLATION
Eman Baghlaf
WHAT IS MACHINE TRANSLATION ?
Machine translation (MT):
The standard name for computerized systems
responsible for the production of translations from
one natural language into another, with or without
human assistance.
Earlier names such as “mechanical translation” and
“automatic translation ” are now rarely used in
English; but their equivalents in other languages are
still common like in French and Russian languages.
• The term does not include computer-based
translation tools which support translators by
providing access to dictionaries and remote
terminology databases, facilitating the transmission
and reception of machine-readable texts or
interacting with word processing, text editing or
printing equipment.
• It does, however, include systems in which
translators or other users assist computers in the
production or translations , including various
combinations of text preparation, on-line
interactions and subsequent revisions of output.
The output of most MT systems is revised (postedited). In this respect, MT output is treated no
differently than the output of most human
translators which is normally revised by another
translator. However, the types of errors produced by
MT systems do differ from those of human
translators.
While post-editing is the norm, there are certain
circumstances when MT output maybe left
unedited(as a raw translation)or only lightly
corrected. Output may also server as a rough draft
for human translators , as a pre-translation.
Computerized
Approaches
Machine Translation
(MT)
Machine-Aided
Human Translation
(MAHT)
Human-Aided
Machine Translation
(HAMT)
Types of computerized approaches:
Machine Translation (MT)
Any system that actually performs a translation.
Machine-Aided Human Translation (MAHT)
They are systems which perform the task of translation
but rely on the intervention of human translator at
various stages in the translation process.
• Human-Aided Machine Translation (HAMT)
• Systems such as word processors, dictionary
management tools, TERM BANKS, and various look-up
facilities which support the translator but do not actually
perform the translation task.
•
•
•
•
•
HISTORY OF MT
• Although the idea of mechanized translation via an
intermediate universal language has been around
since the seventeenth century.
• The first concrete proposal for a translation machine
can be dated precisely by the simultaneous ,but
unconnected issue of patents in 1933 to the Russian
Petr Smirnov-Troyanskii and to the Armenian
Frenchman Georges Artsrouni.
THE AIMS OF MT
The programs can produce raw translations of texts
in relatively well-defined subject domains, which
can be revised to give good –quality translated
texts at an economically viable rate or which in
their unedited state can be read and understood
by specialists in the subject for information purposes.
• In some cases, with appropriate controls on the
language of the input texts, translations can be
produced automatically that are of higher quality
needing little or no revision.
Researchers and developers of MT systems can
ultimately aspire only to producing translations
which are useful in particular situations.
MT is part of a wider sphere of pure research in
computer-based natural language processing in
computational linguistics and artificial intelligence
which explore the basic mechanisms of language
and mind by modelling and simulation in computer
programs.
Research on MT is related to these efforts, adopting
and applying both theoretical perspectives and
operational techniques to translation processes,
and in turn offering insights and solutions from its
particular problems.
•
•
•
•
•
•
•
•
•
•
Classification of MT according to different languages:
1-Bilingual systems
They are designed for one particular pair of translation
2-Multilingual systems
They are for more than two languages
Classification of MT according to directions:
Uni-directional systems
In one direction only
Bi-directional
In both directions
MT ACCORDING TO ITS USERS
Readeroriented MT
Writeroriented MT
• Classification of machine translation according to its
users
A. Reader-oriented MT
• The term is used by Sager(1994) to refer to the speedy
production of TL version of a text by means of machine
translation in order to inform TL reader of the content of
SL.
• The result of such a procedure is raw output or in other
words artificial product which may require more reading
and effort than a human translation because it has not
been post- edited.
• *TL: target language / SL: source language
• It has the advantage of being available quickly
more than a polished version, and can be relied on
to contain relatively small number of lexical and
terminological mistakes.
• Sager points out that such translation develop the
ability to read the artificial machine-produced
language which they typically have a high degree
of fluency.
• An example of that is the system SYSTRAN, which is
used by the US air Force to help survey scientific
and technical literature.
B. Writer –oriented MT
• It involves a writer in pre-editing process in which
the machine asks questions about elements which
the machine cannot analyze (for example because
they are not included in its grammar or
dictionaries).
• The main advantage of this procedure is that the
writer’s original intention will be translated by the
machine, rather than translator’s interpretation of
what the writer wished to communicate.