Preparing Data for the Agent

To develop a dialogue design more easily and quickly, it is helpful to have a history of conversations with the customers. For example, when creating an AI for technical support, phone call records with support operators can be useful. Voice recordings are transcribed and clustered using mining tools. As a result, the dialogue history is divided into topics, and for each topic, there are examples of

@Training Phrase⁠

and ready-made responses from the operator. Examples of

@Training Phrase⁠

are later included in the

@Training Dataset⁠

@Intent⁠

s, and the operator's responses are used when writing AI responses.

When such historical data is not available, there are several ways to populate the AI with content:

Provide a detailed description of the business process and come up with the data yourself.

Find publicly available chatbots for automating similar business processes and adapt the dialogue scenarios used in them to your needs.

Consult with experts at Graphlogic.ai for advice on your case. We can prepare a simple test AI right away if you provide some data. Details can be found in the article

here⁠

To create high-quality training data for the NLU model of your AI, refer to

this article⁠

for guidance.

Want to print your doc?
This is not the way.

Try clicking the ⋯ next to your doc name or using a keyboard shortcut (

CtrlP

) instead.