Concepts

An intent defines an end-user's intention for a conversation. An intent contains phrases, parameter, responses, entities, scripts, macros and documents.

Phrases are sample phrases of what end-users might ask. When the end-user input is similar to one of the phrases, IntelliService triggers the intent. The machine learning capability of the conversational AI automatically trains possible conversations and expands phrases list with similar phrases. Also, the phrases are marked with entity parameters.

In the Phrases, specific word or value defines direct entity value. These words or values are highlighted as parameters to extract values from specific parts of the end-user input. Highlight the word or value on the phrases to extract them as entity value.

Entity is type of data to requested from user during the conversation to process intents. Entity data can be user input, specific part of user conversation, actionable items like button clicks, selecting options and so on. Entity value can be extracted from user phrases or directly received in the chat conversation from user. For the password reset flow, username and login ID are required to process the password reset request. Chatbot can directly request the user to input the username and login ID details. These details will be extracted as entity value from the user conversation.

Flow is action based intent or workflow based intent. You can define the flow with training intents, entities and scripts to define logical flow of sequence of actions. Apart from the conversational flows, you can develop action based flow using JS script and macros to define the logical execution sequence for the flow.