Hi,
I'm currently working on chatbots. For the theoretical foundation, I was wondering if there exist any frameworks to classify possible types of chatbots. I already found some classification frameworks provided by IBM & Co. But, do you know others which are not provided by the big players? It would be very helpful. Thanks a lot!
- S
Chatbot text classification approaches
- Pattern matchers
- Algorithms
- Neural networks
https://medium.com/@gk_/how-chat-bots-work-dfff656a35e2
https://chatbotsmagazine.com/contextual-chat-bots-with-tensorflow-4391749d0077
Interesting, could you share the types by IBM? Basically there are only two classes - task-oriented and non-task-oriented ones. But frankly speaking there might be not such thing as "non-task" if we think about small talk's social functions for example.
Of course: https://www.ibm.com/blogs/watson/2017/12/3-types-of-business-chatbots-you-can-build/
But I am still not convinced of the classification approaches that I found. ;)
You probably mean not "chatbot classification" but intent classification for goal oriented agents. Or what did you mean on " classify possible types of chatbots " ?
Perhaps "intent classfication" would fit better. I'm not sure which criterion should be used to classify chatbots at the moment. There are classifications that classify chatbots based on the technique used. Then others that look at the functionality of a chatbot (e.g., only for support). Further, there are others that look at character traits (e. g. chatty, supporter, etc.).
Btw, I found another description of types of chatbot (cf. http://us.arvato.com/blog/types-of-chatbots/).
Cause' there is no science around chatbots, all terminology is pretty vague, not to mention classification. However the article you have cited is a nice try
- Badges
- Science topic
- Similar topics
- Data Mining
- Classification