Terms and definitions

Here are some terms and definitions related to the NLU service.

Agent

The agent represents all the components that apply natural language understanding capabilities in the NLU service.

Entity

Entities represent data that can be extracted from user sentences and can correspond to different types: physical entities such as locations and people, and other entities such as date, time, and monetary values. System entities are those that use pre-trained models, meaning the application developer does not need to train these entities. Custom entities are those defined by the application developer and possibly trained.

Intent

In a user sentence within a task-oriented conversational system, intent corresponds to the action requested by the user.

NLU

Natural Language Understanding is an area of Natural Language Processing (NLP), which seeks to transform human language into a structured representation, more appropriate to be used by a machine.