Robotics Reference
In-Depth Information
ALICE:
I understand. What makes you think I am
not from NY?
Judge 4:
All the new yorkers I know seem to have
strong opinions about the Yankees
Pwrhaps
11
there are a few exceptions.
ALICE:
Judge 4:
So you are a new yorker who is indifferent
to the Yankees?
ALICE:
Interesting deduction. Thanks and you are
a delightful person too. Am I a new yorker
who is indifferent to the Yankees? Yes.
Judge 4:
What else are you indiffernent to?
ALICE:
Picture me wearing whatever you imagine.
Judge 4:
I'm afraid you may have misunderstood me.
I was just trying to start a conversation.
ALICE:
Not literally afraid, right? You believe that?
YOU? Can what? Why do you want to do
it so much?
Machine Translation Since 1954
Much of the early work in Machine Translation was based on the some-
what simplistic view that all one needed to do to translate from one lan-
guage to another was to look up word-for-word translations in a dictio-
nary and apply the rules necessary to achieve correct word ordering in
the target language. But this approach took no account of ambiguity
and some other features of natural language and, as one might expect
(especially with hindsight), it resulted in poor translations.
Following the 1954 demonstration of the Georgetown/IBM system,
and with the benefit of Noam Chmosky's work on grammars, published
three years
later, Machine Translation researchers became over-
11
The mis-spelling of “perhaps” here was deliberate. In order to try to appear humanlike most of
the programmers entering the Loebner competition simulate humanlike spelling and grammatical
mistakes, such as pressing an adjacent key on the keyboard (in this case “w” instead of the correct
“e”).
