Machine Translation Process

Machine translation is the process of translating documents without further human intervention. Also referred to as MT, the translation is a procedure where a computer program studies a source text and produces a target text. Using computers to do translation in different language is highly recommended in today's fast pace world in certain applications though it also has its own limitations. In some cases machine translation is preferred while in other cases human translation is more preferable. However, we cannot say that machine translation is completely out of human intervention. Human intervention is involved in the form of pre-editing and post-editing.


History of Machine Translation
The history of machine translation started way back in the 1950s. The first work of translation was published in 1954 in the Georgetown experiment involving fully automatic translation of more than60 Russian sentences into English. The experiment was a great success and the authors claimed that machine translation would be used in translations within three or five years. However, the real progress was very slow. The ALPAC (Automatic Language Processing Advisory Committee) report in 1966 further reduced the investment in Machine translation because the report evaluated the progress in computational linguistics in general and machine translation in particular and was very skeptical to research done in machine translation so far and gave more emphasis to the need for basic research in computational linguistics. However, starting in the late 1970s and beginning 1980s, with the impact of personal computer revolution, with the increase in computational power, more interest began to be shown in statistical models for machine translation. There was growth in the use of machine translation as a result of the beginning of less expensive and more powerful computers. With the 1990s, the importance of machine translation further increased (for better or worse) and the use of "translation engines" on the Internet to allow for translation of websites and email languages. Today there are many software programs several of them online for translating source language. Such software includes the SYSTRAN system which powers both Google translate, AltaVista's Babelfish, StarDict etc. These tools produce a rough translation that gives the summary of the source text.

Recent Research
The area of machine translation has in recent tears has seen major changes. Presently a large amount of research is done into example-based machine translation and statistical machine translation. Today, few companies make use of statistical machine translation commercially. For example:

• Language Weaver: Sells translation services and products.
  
  
• Google: Make use of statistical machine tool system for some language combination in Google's language tools and
  
  
• Microsoft: Make use of MT system to translate knowledge based articles.

Advantages of Machine Translations
Some of the advantages of machine translations are as follows:

• Machine translations work at a faster rate than human translations.
  
  
• The second advantage of machine translation is that it gives drop dead consistency. This implies that there is nothing to worry about someone taking too much creative license with your translation or a translator forgetting how a particular word was translated in the previous pages. MT will translate that particular word in the same way. However the drawback is MT will show the same error over and over again unless and until it is asked not to.
  
  
• Another advantage of machine translation is that it is comparatively cheaper. It is one time cost -the cost of the tool and its installation.

Strategies Involved in Machine Translation

• The direct strategy: The first to be used in machine translation systems is the direct strategy, which involves a minimum of linguistic theory. It is based on a predefined source language-target language binomial in which each word of the source language is linked to a corresponding unit in the target language. However here there is a unidirectional correlation, for example from English to Spanish but not the other way round. The following is an example that shows the direct translation strategy:
  
  
  
  
• The transfer strategy: The transfer strategy gives importance on the concept of "level of representation". This strategy involves three stages. The analysis stage, which is the direct strategy, describes the source document linguistically, using a source language dictionary. The transfer stage changes the results of the analysis stage and produces the linguistic and structural equivalents between the two languages. A bilingual dictionary is used to translate the source language to target language. The third stage is the generation stage which produces the target language document based on linguistic data of the source language by using a target language dictionary.
  
  
• The pivot language strategy: This strategy is based on the idea of creating a representation of the document or text independent of any particular language. This representation functions as a neutral, universal translated text that is different from both the source language and the target language.

Where does machine translation work well ?
The success of machine translation is seen in two kinds of documents.

• The texts which have very concrete subjects and follows a regular procedure. There is no requirement of understanding the content to translate the text. The technical handbooks belong to this group.
  
  
• The other documents are those where the readers need a rough translation. The texts that are rarely published and mainly communicative in multinational and national companies belong to the second group.

Besides the above, the machine translation works well on the Internet. Basically, the machine translation is appropriate for the following:
Generally it can be stated, that machine methods can be appropriate:

• For technical/informative texts.
• For texts consisting of many repetitions.
• For texts where it is necessary to keep consistency.
• For texts with small vocabulary.
• For texts which need to be completed quickly.