Page 12 - Міністерство освіти і науки України

The process of disambiguation is partly based on logic but the larger part of it is
what we call intuition.
Now, let us sum up by saying: “We understand people speaking the same
language, we can read, write, and translate due to continuous analysis of the context and
situation and the appropriate use of the relevant background information”.
As for computers they have no sense of intuition which could help them in the
process of disambiguation. That`s why only on some texts, particularly highly technical
texts treating a very narrow topic in a rather dry and monotonous style, computers
sometimes do quite well.

III Basic machine translation methods

The first stage of computer translation development was characteristic of the so
called “encoding – decoding” approach. This approach, which still remains one of the
basic methods of machine translation is usually called the direct or icon method. The
method is based on establishing a direct relationship between the source and target
dictionary entries. The target entries are regarded as regular counterparts of the source
ones. As a result we have a direct word-for-word translation and more often such a non-
grammatical translation makes no sense and, therefore, it is to be somehow rearranged
and smoothed. To improve the quality of direct translation the two following methods
are usually applied: syntactic filters and statistical ranking of translation equivalents to
select the most probable ones for the subject matter discussed in a particular text being
translated.
The second method of machine translation is the transfer-based method.
According to this method of translation grammars of the source and target languages are
matched in the process of translation by a set of rules called transfer.
In a transfer-based system the process of translation comprises the following
processing steps:
A. Morphological analysis – word forms are identified with the dictionary
entries.
B. Syntactic analysis – syntactic representation of the source text is formed by
the syntactic analyzer (called parser).The information is passed over to the
transfer module.
C. Transfer – the transfer module receives the syntactic representation of the
source text and converts it into an intermediate representation .
D. Syntactic synthesis – a final representation is formed by matching the
transferred structures.
E. Morphological synthesis – at this stage the target text is obtained.
Transfer-based systems rather often comprise a semantic component the purpose
of which is to improve the accuracy of translation .
The third basic method of machine translation is pivot language-based machine
translation. In a way it is similar to the transfer-based; however there are some
important differences.

10