0

I don't understand why Google Translate translates the same text in different ways.

Here is the Wikipedia page of the 1973 film "Enter the Dragon". You can see that its traditional Chinese title is: 龍爭虎鬥. Google translates this as "Dragons fight".

Then, if we go to Chinese Wikipedia page of this film, and search for 龍爭虎鬥 using Ctrl-F, it will be found on several places:

enter image description here

But if we try to copy the hyperlink of Chinese page into Google translate, it will be the word "tiger" from somewhere:

enter image description here

Even more, if we try to translate Chinese page into English using build-in Chrome translate, it will be sometimes translated as "Enter the Dragon", in English manner:

enter image description here

Why it gives different translations for the same Chinese text here?

nbro
  • 39,006
  • 12
  • 98
  • 176
john c. j.
  • 111
  • 6

4 Answers4

5

It's not quite clear what you are asking. So I'll answer in separate parts.

Why is the translation different from the official title?

It could be simply because machine translation is not perfect, or our human translator took some creative liberties when translating. In this case it seems to be both.

Note that 龍爭虎鬥 properly translated doesn't mean either "Dragons fight" or "Enter the dragon". Literally translated, it means "dragon compete tiger fight". It belongs a family of well-formed idioms called "Chengyu", which describes a situation where there is fierce fighting or competition.

So you can see that neither translations fit.

Why does Google Translate give me different translations on the same phrase?

Context matters when we read! So translating a phrase in isolation doesn't guarantee that the same phrase has the same meaning/translation in all other parts of the text.

For example, green is a color, but the word "green" can also be used as like: "Alice is green with envy" or "Bob has green thumbs", of which neither instances of the word "green" refers to the color.

Considering the technical side of things, Google translate probably uses some kind of RNN in its pipeline. RNN's are influenced by past states, meaning that what it outputs now as a function of what it reads in is dependent on the RNN's past state. Which is similar to the issue addressed above.

k.c. sayz 'k.c sayz'
  • 2,061
  • 10
  • 26
2

As you know google translation works base on statistical methods. In statistical translation, many parameters can be related to the final result. One of these parameters is co-occurrence of words in a sentence. Hence, as this translator learn languages from different utterances by human and pre-written translations, and different parameters in the text are involved in this learning, It could be possible one word has a different meaning in different texts.

OmG
  • 1,731
  • 10
  • 19
1

Chinese words 'fu' means with different intonation marks either happiness, huspand or tiger. Without correct intonation notation in English it may translate in Chinese as Tiger.

Movie title has Chinese 'happiness' character, but Google mixes it as Tiger.

mico
  • 927
  • 1
  • 8
  • 16
1

Google uses user input to improve translation. Some user may have provided an input to the Traditional Chinese characters using English characters only instead of pinyin, which would introduce a mistake into the data used by the translator. Since the model is statistics based, such a mistaken translation can only be assigned a lower but non-zero probability of correctness, but to erase it from the system entirely you would probably have to do that by hand or by introducing some general rule (e.g. if the probability is so small that it can not be stored in a variable of a given size, round it to zero).

Jan Kubát
  • 11
  • 2