How to train a bot in Mandarin?

Hi everyone, I am posting in this discussion forum for the first time, so please ignore my mistake if I can’t describe my problem properly. I found many articles to train the bot in the Chinese language, many of them are referring to use the total_word_feature_extractor_chi.dat to train the bot.

  • My first query here, how to get the data from the site [https://pan.baidu.com/s/1kNENvlHLYWZIddmtWJ7Pdg](https://baidu) if I don’t have a Baidu account and I am from India.
  • Also, I will be glad if I will be provided with the latest official documentation to train the bot in the Mandarin language.

Hello @neoaman and welcome to the forum! Thanks for your question.

You can use a pre-trained language model for that purpose. We have documentation here on how to configure this. TLDR, you can use either spaCy or MITIE for that purposed, both should support Mandarin :slight_smile: