This article covers key vocabulary to help you get started on the journey of customer service automation.
The key vocabulary terms to familiarize yourself with include:
AI Model
An AI (artificial intelligence) model is a program that has been trained on a set of data to recognize certain types of patterns. The AI model is the brain of the AI agent, it sees an expression, decides which intent it should go to, then provides the reply that they build. Learn more.
Expressions
An expression is a customer message that has been trained to a specific intent. An intent should be trained with a mixture of expressions that reflect the data and customer behavior. The same expression or similar expressions should never be trained to different intents. Learn more.
Intent
An intent is a grouping of semantically-similar messages, or expressions that represents a customer query. These are triggered by understood user messages. Learn more.
There are two categories of intents:
- Meaningful: represent the most frequent customer queries identified in your data ie: order status, order cancellation, return policy
- Structural: also known as Small Talk, support the conversation flow and are not AI agent or industry specific ie: greetings, affirmative, I want to speak with a human.
Natural Language Processing (NLP)
Also known as Natural Language Understanding (NLU). NLP is an AI technology that can learn with ample examples and adapt to different sentence structures and spellings, such as synonyms and typos. This is particularly important in customer service as people pay less attention to correct capitalization and punctuation. Learn more.
System replies
System replies, unlike intent replies, are not used to answer any queries. Instead, depending on the reply, the purpose differs:
- Welcome customers - Welcome Reply
- Inform customers when there are no agents to take the chat - Failed Escalation Reply
- Inform customers that their query is not understood - Default Reply
Template Replies
Template replies are not triggered by user messages. Instead, they are used for drafting, transferring flows, and easy dialogue management. Learn more.