Nowadays, most people are familiar with the term “machine translation”, that is, a translation performed by a computer without human intervention. For example, when we don’t know what a word means in another language, we normally look up its meaning on Google Translate, which is just one of the existing machine translation tools.
While these translation tools can be useful in an emergency, it is also true that the quality of the final text is by no means guaranteed. We will explain why and the type of most common errors, depending on the type of machine translation tool.
How many types of machine translation exist
We can identify and distinguish three types of machine translation:
1. Rule-based machine translation (RBMT): is based on source and target language grammars, bilingual dictionaries, and transfer rules. Its biggest weakness is that it is unable to translate linguistic structures, words or expressions that do not appear in those tools.
2. Statistical Machine Translation (SMT): requires a monolingual corpus in the target language and another parallel corpus with the translation from the source to the target language. Since its function is based on calculating the probability of success in translation, the problem is its huge dependence on the languages in question (whether they are similar or not), the quality of the corpus, the field or specialization, among others.
3. Neural Machine Translation (NMT): is based on large parallel corpora, emulating the way our neurons work by associating words with underlying information to form associations of ideas and therefore be able to translate. The downside is that because it uses characters and sequences, it sometimes generates non-existent or nonsensical words.
Most common errors in machine translation
Translate first names: ability to translate a surname or toponym with low-quality machine translation tools.
False friends: If the corpus is not of good quality, the automatic system can sometimes use inaccurate translations, since it does not know how a word is normally translated. An example is the Spanish word “carrera”, which in some cases refers to “academic degree”, but which some tools translate as “career”, thus changing its meaning.
Word order: An automatic translation system can make errors when interpreting a sentence. For example, usually in English a verb that ends in “-ing” when following a preposition has nothing to do with a continuous tense verb form, yet this case is unlikely to be recognized by an automatic translator.
Use of incorrect meanings in the case of polysemy: This can happen when working with low-quality corpus, as the system does not detect the context and uses an incorrect meaning in the dictionary.
Inability to detect and translate “new” terms: if a word does not exist in its lexicon, the automatic tool will not be able to translate it. The same also happens in the case of neologisms or puns.
Errors due to homographs: automatic systems hardly distinguish two words written in the same way, but which have different meanings or even different grammatical categories.
Uso scorretto delle lettere maiuscole/minuscole: alcuni termini sono scritti in maiuscolo, a seconda del contesto; contesto che spesso un traduttore automatico non è in grado di riconoscere.
Untranslated acronyms: Some acronyms are known to have an official translation, but the quality of some machine translation tools is insufficient for these to be incorporated into their corpus.
Literal translations: The most rudimentary automatic tools simply translate words, not meanings. So they may end up using structures or words that a native speaker of the target language would never use.
LingoYou, your translation agency
LingoYou is the translation agency that allows organizations around the world to communicate correctly in any language and with any tool.
Marketing Team LingoYou
Send us an email