AUTOMATED TEXT TRANSLATION

Automated translation - is a sub-field of computational linguistics that investigates the use of software to translate text or speech from one language to another. Machine translation can use a method based on linguistic rules, which means that words will be translated in a linguistic way. You can see the scheme of this approach below.

Now we will compare two translators.

YandexTranslator (formerly Yandex.Translation) is a Yandex web service designed to translate a part of a text or a web page into another language.

The system builds its correspondence dictionaries based on the analysis of millions of translated texts. The text for translation is first compared by the computer with the base of words, then with the base of language models, trying to determine the meaning of the expression in context.

In addition, the translator is embedded in the Yandex Browser and automatically offers to translate text in a foreign language.

The translator from Yandex, like other automatic translation tools, has its limitations. This tool aims to help the reader understand the general meaning of the content of a text in a foreign language; it does not provide accurate translations.

Google Translator (Google Translate) - Google's web service, designed to automatically translate part of a text or web page into another language. For some languages, users are offered translation options, for example, for technical terms that should be included in future updates to the translation system.

Google uses its own software. It is assumed that the company uses a self-learning machine translation algorithm. In March 2017, Google fully switched the translation engine to neural networks for better translation.

The translator has 103 languages available.

Google Neural Machine Translation (GNMT) is a neural system (NMT) developed by Google and introduced in November 2016, which uses an artificial neural network to improve the fluency and accuracy of translation.

GNMT improves the quality of translation using the example-based machine translation method (EBMT), in which the system “learns from millions of examples”.

It often happens that we need to use an online translator. Usually, Google and Yandex. Translators are at hand. What are convenient services, what features do they have and which one is better?

When installing an application from the store, each user is interested in the issue of functionality, the presence of a user-friendly interface and stability of work. Of course, products from Google appeared much earlier, and in most cases Yandex simply tries to reproduce ready-made applications in their laboratories, slightly changing them.

Sometimes this kind of developer behavior may seem demeaning, but in this case, the global race for technology is worth it.

Yandex.Translate can be called a worthy and high-quality competitor to Google Translate, because it does an excellent job with its main function. Well, д the developers add a few additional functions, it will be able to become a leader among similar programs.

During our research we have developed our own application. This application was written on Java programming language using regular expressions. The method is that a file with words in English and their translation into another language is loaded into the application. With the help of regular expressions in the file there is the word and the next character is /. After that, when you press the Translate

References

1.        https://en.wikipedia.org/wiki/Machine_translation

2.        https://en.wikipedia.org/wiki/Translation

3.        https://en.wikipedia.org/wiki/Google_Translate#Functions

4.        https://en.wikipedia.org/wiki/Yandex.Translate#Translation_methodology