ACL4 Machine Translation Exam 1995
Andy Way
Dublin City University
Attempt five questions.
Q 1.
Critically evaluate the following comments from Arnold et al,
1994:7:
- MT is a waste of time because you will never make a machine that can
translate Shakespeare.
- There was/is an MT system which translated The spirit
is willing, but the flesh is weak into the Russian equivalent of
The vodka is good, but the steak is lousy, and hydraulic
ram into the French equivalent of water goat. MT is useless.
- Generally, the quality of translation you can get from
an MT system is very low. This makes them
useless in practice.
- MT threatens the jobs of translators.
- The Japanese have developed a system that you can talk to on the
phone. It translates what you say into Japanese, and translates the
other speaker's replies into English.
- There is an amazing South American Indian language
with a structure of such logical perfection that it solves the
problem of designing MT systems.
- MT systems are machines, and buying an MT system should
be very much like buying a car.
Q 2.
Discuss some of the benefits that MT can bring to a translation
bureau? How can users of MT systems help the system to improve both
its throughput and quality of output? How would this be
achieved with particular systems you have practical experience of?
Q 3.
Outline some of the issues involved in the processes of Analysis
and Generation? How do their demands differ?
Q 4.
Show how (a) a transfer-based system and (b) an
interlingual system would translate the sentence ``Each political
party depends on the loyalty of its supporters'' into a language of
your choice, commenting on how your chosen representation schema deals
with the issues involved. What are the advantages and disadvantages of
such systems?
Q 5.
How would you integrate a morphological component into the
dictionaries of an MT system? Does this depend at all on the type of
MT system being used?
Q 6.
Comment on some of the similarities and disparities of translating
between two human languages as opposed to compiling between two
programming languages, focussing in particular on
implementation issues.
Q 7.
Discuss, with reference to at least one system you have experience of, the
advantages of taking a sublanguage approach to MT.
Q 8.
What issues are involved in the evaluation of MT systems?
Can we compare MT using the same criteria as those used to evaluate
translation as performed by humans?
Andy Way, 10th March 2000.