A model begins with a list of general user intentions, called intents, such as Book Flight or Contact Help Desk. You provide an example text, called an example utterance, for the intents. Then, you mark the significant words or phrases in the utterance, which are called entities. A model includes the following:
- Intents: Categories of user intentions (the intended action or result)
- Entities: Specific types of data in utterances, such as numbers, emails, or names
- Example utterances: Example text that a user enters in the client application
Some examples of utterance and response are shown here:
Logic Apps has the following two actions as built-in connectors:
The following are the APIs that are part of LUIS:
|
API name |
Explanation |
|
Get entity by type |
For a given entity type, the operation returns the best matching entity model from the LUIS prediction object. |
|
Get prediction |
Given some input text, this operation returns a prediction based on a pre-trained model. The prediction object returned can also be used as input for other LUIS actions. |