BESSER-Bot-Framework
BESSER-Bot-Framework copied to clipboard
BESSER-PEARL
Design and implement chatbots in Python
Similar to #9 in WebSocket Platform
When getting the value of a matched parameter in a state body, e.g. `temperature = predicted_intent.get_parameter('temperature').value` we always obtain a `str` object, therefore, we cannot do `if temperature > 0`,...
- Unit tests - Integration tests - CI/CD (Github Actions)
In case one would like to develop a bot capable of using multiple languages, there should be a language model for each language. This way, if a language switch is...
While working on a bot model, I noticed two things. First of all, each restart of the bot (and thus each re-training of the NLP model) causes the recognition scores...
The current NER only works with exact matches in the entities. Slight variations would miss the entities and therefore carry this problem to the intent classification module (missing both the...
When using following training examples: ```python confirmation_intent = bot.new_intent('confirmation_intent', [ 'yes', 'yeah', 'of course', 'i am happy to say yes' ]) rejection_intent = bot.new_intent('rejection_intent', [ 'no', 'nah', 'never', 'no i...
As a reference, see Xatkit's system entities list (it is the list of DialogFlow's system entities) https://github.com/xatkit-bot-platform/xatkit/wiki/Intents-and-Entities#system-entities