I want to build a chatbot mostly BERT(Transformer) based in the German Language. But I do not find any German chatbot data set!

So does it make sense to use google translator API to translate the English dataset to German and then train the model on it?

Any idea where I can find German datasets or solve this issue?

  • $\begingroup$ which one shall I delete? Where are we more likely to find answer? $\endgroup$
    – Bot_Start
    May 30 '20 at 12:14
  • $\begingroup$ It doesn't allow me to delete so I deleted the other one. So I hope I get more answers on this itself. $\endgroup$
    – Bot_Start
    May 30 '20 at 12:32

To start with chatbot first of all you to decide which type of chatbot are you trying to build.

  1. Order placing bot
  2. Chitchat bot

Both required different approach to solve the problem.

Order Placing Bot required multiple model for different task like intent identification, named entity recognition, state machine.

Chitchat bot required only 2 person conversation dataset which is available easily on kaggle.com

But if you are looking for specific language dataset then it difficult to find it in both type of bots. For that either you use any translation api which you to pay for it or use web scrapping techniques to do same task at free of cost.

  • $\begingroup$ Ok, here the intention is to first pick the SOTA chatbots of English and then fine-tune/ train it on the German data set. But after a lot of searches, I couldn't find any data set in the German language. Hence the question. Also, there is a free and unlimited python google API to translate English to German. pypi.org/project/googletrans. Do you think it makes sense? $\endgroup$
    – Bot_Start
    May 30 '20 at 12:34

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.