I frequently need to translate product names for hundreds of similar products -- and I have a list of past product names. Is it possible to train AI to review past translations and translate? It doesn't have any special grammar, simply the name (with some industry-specific usage that a general machine translator can't do.) What would I need to do to get started?

  • $\begingroup$ Are the product names unique? For eg all past products have different unique words as it's names, while a new product has a new word as it's name....Or are all the products have a combination of some common words as their names? Also what do you mean by machine translator? $\endgroup$ – user9947 Mar 8 '19 at 12:13
  • $\begingroup$ One of the big problems here is if the product names are idiomatic or involve wordplay. This feels like a very difficult NLP problem. By contrast, technical names would be fairly easy. It seems to me that translating product description accurately would be more important than translating product names. $\endgroup$ – DukeZhou Mar 8 '19 at 20:34
  • $\begingroup$ They are unique. i.e. One word in Chinese always translates to one word in English. However, the way the words are ordered and splitting up Chinese text into 'words' require some learning (not a straight-forward find/replace) that's why I thought some kind of pattern-matching and learning can solve this. $\endgroup$ – ytk Mar 22 '19 at 4:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.