56 votes
Accepted

Why don't you get back the original text when you use translation software to translate something into another language and then back into English?

There is no one-to-one correspondence between languages and their vocabularies. This means it is impossible for a computer translator to be invertible. The translator's task going from language A to ...
gaeguri's user avatar
  • 1,485
13 votes
Accepted

How good is a round-trip translation as a machine translation quality evaluation technique?

A so-called round-trip translation is not a reliable signal that the translation worked well, for a few reasons. Noisy parallel corpora are bi-directional. Machine translation systems train on ...
Adam Bittlingmayer's user avatar
10 votes

Did Google Translate supply North Korea with the word "dotard" because of J.R.R. Tolkien's "Lord of the Rings"?

It is unlikely, since Google translates the Korean original 늙다리(미치광이) (neulgdalimichigwang-i) as "an old man lunatic", where "늙다리" (neulgdali) conveys "old" and "미치광이" (michigwang-i) conveys "maniac". ...
user6726's user avatar
  • 83k
9 votes
Accepted

"Episode 78" machine mistranslation etymology

The source of this mistranslation series was identified by Japanese internet users as parallel corpus data contaminated by a Japanese TV reality show program あいのり Ainori. According to this article, a ...
broccoli forest's user avatar
8 votes
Accepted

Some “linguistic formulas” to translate French texts into English?

Unfortunately, it can't be done. Translation is an immensely complicated process, and nobody's ever made a mathematical procedure that can do it with any reliability. The best we have right now are ...
Draconis's user avatar
  • 65.2k
7 votes

Does machine learning still need a better linguistic theory?

This is a funny question because of 'machine learning' (ML), 'still', and 'better'. Presumably you mean 'machine learning methods in NLP' (Natural Language Processing), because I'm having a hard time ...
Mitch's user avatar
  • 4,445
6 votes

Accuracy of automatic (machine) translation of two languages in same language family

Update: Much has changed since 2019. I added updates at the bottom. You're right to suspect that the accuracy can potentially be very good, but, in practice, unfortunately, as of 2019, most of the ...
Adam Bittlingmayer's user avatar
6 votes

What language pairs are underserved, and would most benefit from machine-learned bilingual tools?

This question is very important and possible to answer empirically, however, words and concepts do not map 1:1 across languages so the mentioned assumption that bilingual dictionaries will have a ...
Adam Bittlingmayer's user avatar
5 votes

Is there an English to Coptic translator online?

Machine translation in general is in its infancy. Even for major languages like Mandarin Chinese and Spanish, computers have trouble with context-dependent concepts such as verb inflection and words ...
Locoluis's user avatar
  • 456
5 votes

Are European Union parallel multilingual texts ideal for machine learning of machine translation?

Europarl is a classic corpus for research papers, used at the main conference - WMT - and by some of the top people in the field. It would be useful for training a translation system specifically for ...
Adam Bittlingmayer's user avatar
5 votes

Some “linguistic formulas” to translate French texts into English?

Your examples are of cognates and loanwords (mostly loanwords), which means words that have the same origin in English and French, or that English borrowed from French. English does have many words ...
LjL's user avatar
  • 1,849
5 votes
Accepted

Norwegian Translation Codes (no, nn, nb) - Which to use on a website?

I assume your concern is with regard to Norwegians and not compliance with some statutory requirement (if there is any such requirement, which I doubt, I am certain that it wasn't arrived at by ...
user6726's user avatar
  • 83k
4 votes
Accepted

Which original language would cause "your" to be translated to "their" in English?

I see now… what was the reason that they blocked me if I did not have any rechasado would seem to come from (Latin American) Spanish (rechasado not being translated because it should be written with z ...
dumetrulo's user avatar
  • 226
4 votes
Accepted

Noisy channel model explanation

If we take into account only P(D|Ai), the probability is similarly wrong/right as P(Ai|D). Since both are taken from the double-language corpora and thus not really good. But because in the right ...
Septinel's user avatar
4 votes

Some “linguistic formulas” to translate French texts into English?

I would call what you're describing character-level rules-based machine translation (RBMT). RBMT lost out to SMT in the early 2000s, and SMT evolved into NMT in the 2010s. But character mappings are ...
Adam Bittlingmayer's user avatar
3 votes
Accepted

How lexical ambiguity addressed in phrase-based SMT

Well, a word like 'bank' that is ambiguous between 'financial institution' and 'side of a river' will often be disambiguated by context. So, the first sense is talked about near words like 'money', ...
Russell Richie's user avatar
3 votes

Question about Range of BLEU

The BLEU score is between 0 and 1, but is sometime expressed as a percentage, i.e. ranging from 0 to 100%. E.g. http://www.statmt.org/moses/?n=Moses.SupportTools#ntoc5 returns a score between 0 and ...
Franck Dernoncourt's user avatar
3 votes

Does machine learning still need a better linguistic theory?

After reading Mitch's answer and some of his references, I think a prior question needs to be addressed: Does machine learning need a theory? In the essay from Norvig, Chomsky, I don't see any theory ...
Greg Lee's user avatar
  • 12.5k
3 votes
Accepted

Bing Translate isn't same as the previous translation

Translating isn't a one-to-one process. For a simple example, consider the French words si and oui, both of which can correspond to English "yes" in different contexts. Once you've translated them ...
Draconis's user avatar
  • 65.2k
3 votes

Norwegian Translation Codes (no, nn, nb) - Which to use on a website?

As recommended by W3C, you should be using nb as it is more specific than the macrolanguage no. Use macrolanguages with care. Some language subtags have a Scope field set to macrolanguage, ie. this ...
soliz's user avatar
  • 131
2 votes

What are the best NLP sentence alignment tools?

There is hunalign available from github. It is the current standard tool for multilingual aligned corpora.
Sir Cornflakes's user avatar
2 votes

What are the various approaches to detect whether a sentence is complete or not?

For formal English anyway you can parse with spaCy and then iterate through the tagged tokens looking for a finite verb (VerbForm=Fin, as opposed to Ger or Inf). See https://spacy.io/usage/linguistic-...
Adam Bittlingmayer's user avatar
2 votes

Why don't you get back the original text when you use translation software to translate something into another language and then back into English?

There is no one-to-one correspondence between languages and their vocabularies. It is fundamentally different from translating A to B and translating B to A in any machine translation software. For ...
user298438's user avatar
2 votes

Is www.babelfish.com/success/ a rule-based engine?

Without your receiving a response from the contact form, and there being no answer when I call the phone number listed in business directories online, let's see what clues we can find on the website......
Luke Sawczak's user avatar
  • 2,432
2 votes

Is there an English to Coptic translator online?

I know it's late, BUT I did find something else useful in case someone else is searching for this information. It's a bit of a process and only works for single words, though. Go to coptic-dictionary....
Zaneta's user avatar
  • 21
2 votes

Is there an English to Coptic translator online?

As Locoluis explained in his excellent answer, machine translation in general is unreliable. For dead languages, the best attempts are laughable: Google Translate's English-to-Latin, for instance, ...
Draconis's user avatar
  • 65.2k
2 votes

Network of Phonological Relationships

To preface: "similarity" depends on the listener, as well as on the sound itself. Categorical perception is a powerful force. To a native English speaker aspiration is almost imperceptible, while /s/ ...
Draconis's user avatar
  • 65.2k
2 votes
Accepted

Enumerating the possible Pinyin -> Hanzi transcriptions

There are a multitude of online pinyin to Hanzi converters as well as several pinyin-based IMEs that do what you want but not in a way that you can include them in your own program. Over on the ...
acattle's user avatar
  • 2,898
2 votes
Accepted

Will machine translation eclipse the need for human translation some day and if so, by what date do you predict?

Already, Google Translate has reduced the need for knowledge of foreign languages and for translators. There is the possibility that human translators will become much less useful, especial for major ...
user6726's user avatar
  • 83k
2 votes

To which extent language exists disconnected from the physical world?

What we mean by understanding a thing is fitting it into our understanding of the physical laws of our universe, so the answer is: no way. If you could go any way in your understanding of this ...
Greg Lee's user avatar
  • 12.5k

Only top scored, non community-wiki answers of a minimum length are eligible