Some work has been done in the use of multiparallel corporation, i.e. a body of text translated into 3 or more languages. These methods make it possible to use a text translated into two or more languages in combination to allow a more accurate translation into a third language than if only one of these starting languages were used alone. [39] [40] [41] Different programs may work well for different purposes. For example, statistical machine translation (SMT) generally outperforms machine translation for example (EBMT), but the researchers found that EBMT performed better in evaluating translation from English to French. [60] The same concept applies to technical documents which, due to their formal language, can be more easily translated by SMT. Only original works are protected by copyright, so some scientists claim that the results of machine translation are not entitled to copyright protection, as mt does not require creativity. [69] The copyright at issue relates to a derivative work; The author of the original work in the original language does not lose his rights when a work is translated: a translator must have permission to publish a translation. Designated entities must first be identified in the text; If this is not the case, they may be mistranslated as common names, which would most likely not harm the BLUE evaluation of the translation, but would alter the human readability of the text. [36] They can be omitted from the translation of the edition, which would also affect the readability and message of the text.

Interlingual machine translation is an example of rules-based machine translation approaches. In this approach, the starting language, i.e. the text to be translated, is transformed into an interlingual language, i.e. a “linguistically neutral” representation, independent of any language. The target language is then generated from Interlingua. One of the main advantages of this system is to make Interlingua more valuable with the increase in the number of target languages into which it can be converted. However, the only interlingual machine translation system implemented at the commercial level is the KANT system (Nyberg and Mitamura, 1992), intended to translate Caterpillar Technical English (CTE) into other languages. Transliteration includes searching for letters in the target language most likely to match the name in the source language. However, this was mentioned as a deterioration in the quality of the translation at times. [37] For Southern California, the first word must be translated directly, while the second word should be translity.

Machines often transmit both because they have treated them as a single entity. Words like this are difficult for machine translators to process, even those that have a transliteration component….

О сайте