Global Detection Action
This action will parse every text sent to the current interaction by the user. This action will not hold the conversation and wait for an answer but will be triggered anywhere during an interaction flow.
The action uses the following settings:
NLP or Text
Set the type of detection that should be performed
The name of the intents that should be detected. Only global or manually activated intents are listed here.
The minimum detection confidence that need to be reached to trigger this action
Append Detected Entities
If enabled will append detected entities values to the current runtime variables that already hold entity values
A simple text or full Regular Expression match string
The variable that will hold the user text that triggered the action.
This action is not blocking. The conversation is not halted and the user can continue freely. If you want to stop the user until the question requirements have been met you should use input actions.
Text detection can happen in two different ways: using simple text or using a Regular Expression.
- Simple text The text you enter in Control Expression is directly matched with the user input. For example entering this is awesome will match any sentence containing that exact text. You can use a pipe (|) symbol to match multiple sentences. So to match this is awesome and this is great, you can enter this is awesome|this is great in Control Expression.
- Regular Expression Xenioo supports standard Regular Expression notation. The input text will be matched against the regular expression specified in Control Expression.
Regardless or the selected method, text detection is always case insensitive.
If multiple intents are selected, only the best one will be checked against the given confidence.
On User Input Detected. This action will trigger any child operation as soon as the user input is matching the specified Control Expression or the selected intent.
This action produces the following variables after being triggered:
The intent detected during last input
The locale of the intent detected
The full list of all entities values separated by a semicolon
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:
Not Available. Out of AI Detection is not supported inside skills.