Page 13 - Міністерство освіти і науки України

As opposed to transfer procedures which are applied mostly at the syntactic level,
pivot language representation involves all available linguistic information.
Besides, transfer-based translation is intended for a concrete language pair,
whereas pivot language-based systems claim to be universal, i.e. apllicable to any
language.
A pivot language is a formal description of morphological, syntactic and semantic
characteristics of a language unit in the form of one-to-one relationship. Each language
unit is related to a specific invariable atom in the pivot language structure and vice
versa, each atom of the pivot language structure is invariably connected with the units
of various languages.
The fourth basic method of machine translation is called the artificial intelligence
(AI) – based method. The main component of the AI translation model is its so-called
“knowledge base”. According to an AI – based translation model the results of
linguistic analyses at all language levels are verified against extralinguistic information
contained in the knowledge base.
The first three basic methods of machine translation use only context in the
process of disambiguation. In AI – based translation models disambiguation procedures
are radically different and based first and most of all on the analysis of situation and
background information (knowledge base), whereas purely linguistic context analysis
methods serve only as secondary back-up tool.
The knowledge base is the subject thesaurus which contains a specifically
arranged hierarchy of the facts of real world with verbal information playing a
subordinate role as labels for the facts and situations.
Among the basic machine translation methods are also statistical methods.
According to purely statistical methods of computer-based translation it is
presumed that with certain probability each word of the target text may be a translation
of each word of the source text. The various statistical models are very different but
their aim is the same – to estimate probability. For instance, one of the popular models
estimates the probabilities of word order matching in the source and target texts. Other
models estimate the probabilities of word collocations and so on.
It should be noted, however, that none of the machine translation methods appear
in real systems in pure form. Very often it is rather difficult to classify unambiguously
whether one or another system is a typical representative of a particular basic machine
translation method.

IV Conclusions

Computer-based translation is the latest development in modern translation
practice. But unfortunately computer is not capable of providing adequate translation.
Its employment remains restricted in the main to scientific and technological
information and to the sphere of lexicographic work. That is because machine
translation can be performed only on the basis of programmes elaborated by
linguistically trained operators. Besides, preparing programmes for any matter is

11