DeepQA icon indicating copy to clipboard operation
DeepQA copied to clipboard

Published 20 hours ago • verified publisher icon Conchylicultor

Can we work this DeepQA for Chinese dataset?

Open balagopal24 opened this issue 5 years ago • 4 comments

Hello,

I'm trying to include my own chinese dataset by using conversation file <name>.txt, copy it in this repository and launch the program with the option --corpus lightweight --datasetTag <name>

But I could'nt get any responses in chinese. Can anybody help me to solve this issue. Does this chatbot support for chinese language?

Regards, Bala

balagopal24 avatar Aug 25 '18 10:08 balagopal24

Hello Bala, I'm glad to tell you that DeepQA indeed support Chinese. Please prepare you own data for example file named "my_own_data.txt", then put it in dir 'lightweight'. And when you train your model, you should use the command python main.py --corpus lightweight --datasetTag my_own_data. If the command doesn't work, please remove the data in "data/samples" dir. Good luck!

ghostyoona avatar Dec 15 '18 12:12 ghostyoona

good ，show the data cut in word ？

xxllp avatar Dec 17 '18 05:12 xxllp

Hi，Do we need to cut the word if we want to use Chinese? Anyone used with Chinese data successfully?

lhuang9703 avatar May 08 '20 11:05 lhuang9703

I have tried according to what you said, but the answer I got was still in English.Like this: Q: 香港 还 卖 这么 有 爱的 冰棍 ？ A: And the jews killed me.

Zhouziyi828 avatar Mar 10 '22 03:03 Zhouziyi828