In this paper, the overview of machine translation (MT) is presented. The original idea of MT has been investigated since 1950s by many research groups and at present many MT systems have been created and developed around the world. Three approaches of MT systems: direct translation, transfer and interlingual approaches are common systems. The main idea of direct translation approach is word-by-word replacement before the transformation of the structure from source language (SL) to target language (TL). The translation is finished in only one step. The transfer approach transforms the SL to the TL through several stages with different kind of temporarily structural language. Finally, the interlingual approach stands on the idea of transform the SL to the TL through a neutral language or interlingual language (IL).
Language is a tool for communication about the world. Two categories of language are programming language (PL) and natural language (NL). Programming language is a tool for communication between human and computer, whereas natural language is a tool for communication among people. In this paper, we will focus on natural language.
Different nations (or regions) use different languages as a way to communicate to each other. Communication between two people who know two different languages needs an interpreter who understands both languages. Even though most people have an ability to study and understand several languages, it takes time to do so, and so far, nobody is able to understand all the languages in this world. Therefore, the idea of translating languages by a machine, so-called machine translation (MT), was promised in mid-20 century and defined as follows:
"Machine Translation (MT) is the application of computers to the translation of texts from one natural language into another." [Hutchins 1986]
Having a machine translate languages for us is not only by mean of the convenience to our communicating performance, but it is also proven to be very fast. However, differences of each language cause many hart-to-solve problems. Although several languages (most European languages for instance) have many things in common since they are derived from the same predecessor (such as Latin), there exist a great number of differences between them such as the characters and the pronunciation of words, the semantic or the structure of sentences, including the background knowledge of people using the language.
Such differences are a great challenge for scientists and linguists to create an intelligent machine that can understand natural languages.
Nowadays there are many people interested in the machine translation. Searching through world wide web, you can see thousands of web sites involved with the machine translation. Many research groups around the world, such as in Europe, America and Asia, are working on this issue. Also, there are many MT systems produced and being developed...