This action will hold the conversation and validate user input against a Xenioo NLP Intent. The conversation will hold until the intent is detected with enough confidence or if the user, when allowed, skips the question.
The minimum detection confidence that need to be reached to trigger this action
No Detection Reply
This is the text that will be displayed if the user input does not match the control expression
Reply Pick Mode
When using multiple Wrong Input Reply lines, this option controls how replies are chosen.
Max Wrong Answers
The maximum number of wrong inputs the user can send before the question is skipped. Set to 0 (zero) to never give up on wrong inputs.
Never Skip Question
The question cannot be skipped by user. The input will repeat indefinitely until the control expression is matched or the maximum number of wrong inputs is reached
Intent Skip Question
The user can skip the question if his reply matches a specific intent
Text Skip Question
The user can skip the question if his reply matches a specific text or Regular Expression
This action is halting the conversation. The chatbot is halted at the input point and will not proceed unless the conditions specified in the settings are met.
If the maximum number of wrong answers is reached the Successful User Input will still be triggered: you should check the contents of your the fail triggered variable to verify if the input was successful or not.
If multiple intents are selected, only the best one will be checked against the given confidence.
Reply Pick Mode options will pick a wrong reply (if more than one is specified) in the following way:
The reply will be picked randomly from the list of possible replies
The bot will pick the first reply at the first wrong input, then the second and so on. The bot will keep saying the last one when the end of the lis is reached
Similar to list, except that when the end of the list is reached, the bot will restart from the first answer
Successful User Input. This action will trigger any child operation as soon as the user input is matching the specified Control Expression.
This action produces the following variables after being triggered:
The name of the detected intent
The key of the detected intent
The confidence of the last detected intent
The locale of the detected intent
The number of times the user has failed to match the control expression
True if the maximum number of wrong inputs has been reached. Otherwise empty.
Every entity specified in the detected expression will be translated to a runtime variable named after the entity name.
Channel specific availability follows the table below:
This action can be used only inside interactions and not as child of other actions. Any account can use this action regardless of subscription level.