When you read a text, you analyze its contents drawing on your entire life experience, i.e. mostly on the word meanings. Our algorhitm analyzes the text drawing mainly on the grammar and syntax and only partly using semantic analysis.

It is so, because the grammar and syntax are much better defined than the semantic associations everyone starts learning from the very birth. As a consequence of the above, insignificant errors in the grammar of the input text might sometimes lead to the complete loss of understandability in the translation. A similar effect can be observed if the intended meaning of a given word in the source sentence is not present in the database.

We are constantly adding knowledge to our databases, so every application update should give you an improvement of the translation quality.