Yandex. Translation: optimization and use

Hello. Today I would like to touch on the topic of machine translation. When I got a job in a small IT company immediately after graduating from university, I noticed that many novice IT specialists who were not fluent in English often used various web services to work: Yandex.Translate, Google Translator, etc. I won’t argue that these services may be partially useful for a general understanding of the text, but for inexperienced users to get adequate translation using them is very difficult. Sometimes these translations simply call “Hochma”. I remember once one newbie in the simple sentence "I have a wife and two young children" received the following translation: "I have a wife and two young children." By the way, the translation did not bother him at all.

So, is it possible to achieve adequate translation using electronic translators? Of course you can, but for this you need to be able to optimize them for your tasks. Today, I propose to consider a small example of how this can be achieved using the Yandex.Translate web service. An example would be useful to those whose level of English is extremely low, while they want to learn how to read technical literature. If this topic is useful to someone, it can be considered in more detail.

So, what to do if you know English very badly, and you need to read any technical literature in a foreign language? There is nothing wrong with that, you can use electronic translators to solve this goal, optimizing the translation results. Unfortunately, at the present time computers do not have creative abilities, most of the work done on it requires processing by a person. This provision applies to transfers.

Natural languages, including English, are similar to programming languages, any language is a system of codes. Having studied the syntactic, lexical chains and the stylistics of the English and Russian languages, you can process machine translation to a completely normal, readable version. Yes, it will not be a professional translation, but we do not need this. We just need to get a translation of technical literature from English. Today we look at one of these structures. Having studied all the basic constructions, one can literally half a year later learn how to successfully optimize translations.
')
Design:
(A, an) Noun + is + (A, an) Noun: singular
A cat is an animal.

where noun is a singular noun
A, an = one - indefinite article (meaning one item)
is is a singular verb

Comment: a noun means an object (literally, this is all that we see in front of us, for example, a chair, a table, etc., answers the question who? What?). In English, in front of a noun, to show, for example, that the chair is a single indefinite article a, an (an - before the vowels a, e, i, o, u, a - before the consonants).

Translation: to translate a sentence of this type from English to Russian, you need to remove grammatical elements that are not in Russian. For example, A cat is an animal. We remove the articles and the verb 'to be' (is), we get the translation: Cat - an animal. To translate from Russian into English, you need to add the opposite grammatical elements that are in English, and that is, articles and the verb 'to be' (is).

Now we are looking at how the Yandex.Translate offer of this type of service translates to us:
rus-eng A cat is an animal. Translation: The cat is an animal. Thus, the translation is not entirely correct.
English-Russian Google is a search engine. Translation: Google is a search engine. A cat is an animal. Translation: A cat is a beast.

Thus, it can be said that the Yandex.Translate service translates more or less adequately from English to Russian, although there are lexical and stylistic mistakes. The situation is more complicated with translation from Russian to English, most often Yandex.Translating in this case makes mistakes. We considered only a small example of the translation of the simplest construction of the English language, but even in this case, the translation may often be incorrect when using electronic translators, but knowledge of the basic syntax, lexical chains, terminology and style can help optimize machine translation to a completely “readable” states ".

PS Here was presented only a small example of optimization of translation of one of the simplest structures. The topic is very extensive, and if someone is interested, it can be considered in more detail.