-
Notifications
You must be signed in to change notification settings - Fork 4.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Is there a way to get the accuracy of this chatterBot #459
Comments
ChatterBot uses logic adapters, which are essentially just modules that take input and return a response. Each response that is returned has a confidence value associated with it. The confidence value is a numeric indicator of how accurate the logic adapter thinks the response is. A confidence score is kind of a metric for accuracy but it probably better reflects the breadth of a particular bot's knowledge than the accuracy of it's responses. Arguably a chat bot that responds more confidently may also be more accurate, but that is based on the assumption that the set of logic adapters the bot uses are all completely accurate in their ability to judge the precision of the response they generate. Because ChatterBot's logic adapters each encompass a modular process for selecting a response, there isn't necessarily a common way to gauge the accuracy of all of them. On an individual basis this might be possible but I don't have anything in the code base right now for doing a full evaluation of accuracy. Please let me know if I can clarify anything. |
@gunthercox how do I get the confidence level from |
Currently |
Cool. I am using a quick and dirty method now, I'm not sure if it was meant to be used this way. from chatterbot.conversation import Statement
# initialization
...
statement = Statement(message)
confidence_level = bot.logic.process(statement)[0]
response = bot.logic.process(statement)[1].text |
This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs. |
In Machine Learning area, there always a test data set used for get the accuracy of the model, so, is there any similiar thing for ChatterBot?
The text was updated successfully, but these errors were encountered: