ixp
latest
false
Important :
Communications Mining is now part of UiPath IXP. Check the Introduction in the Overview Guide for more details.
UiPath logo, featuring letters U and I in white

Communications Mining user guide

Last updated Aug 1, 2025

Labels (predictions, confidence levels, label hierarchy, and label sentiment)

A label is a structured summary of an intent or concept expressed within a message. Multiple labels often summarize a message, which means that a label is not a mutually exclusive classification of the message.

As an example, in a dataset monitoring the customer experience, we might create a label called Incorrect Invoice Notification, which describes when a customer informs the business that they have received what they believe is an incorrect invoice.

Label creation and editing actions are primarily performed in the Explore and Discover pages.

Comparing pinned and predicted

You initially create labels by applying one to a relevant message. You can continue to apply them to build up training examples for the model, and the platform will then start to automatically predict the label across the dataset where it is relevant.

A label that you applied to a message is considered pinned, whereas the labels that the platform assigns to messages are known as label predictions.

To learn about reviewed and unreviewed messages, check Annotated and unannotated messages.

Confidence levels

When the platform predicts whether a label applies to a message that has not been reviewed by a user, it provides a confidence level percentage % for that label prediction. The higher the confidence level, the more confident the platform is that the label applies.



The confidence level that the platform has in the predicted labels shades the labels. The more opaque the label, the higher the confidence of the platform is that the label applies.

Label hierarchy

You can organize labels in a hierarchical structure to help you organize and train new concepts more quickly.

This hierarchy can be in the following format: [Parent label] > [Branch label 1] > [Branch label n] > [Child label]

A label can be a standalone parent label or have branch and child labels, which the greater-than sign > separates, that form subsets of the previous labels in the hierarchy.

Any time a child label or branch label is pinned or predicted, the model considers the previous levels in the hierarchy to have been pinned or predicted as well. Predictions for parent labels will typically have higher confidence levels than the lower levels of the hierarchy, as they are often easier to identify.

For more details about label hierarchies, check Taxonomies.

Label sentiment

For datasets with sentiment analysis enabled, a green or red colour indicates a positive or negative sentiment for every label, both pinned and predicted.

Different levels of a label hierarchy can have different sentiment predictions. For example, a review could be overall positive about a Property, but could be negative about the Property > Location.



  • Comparing pinned and predicted
  • Confidence levels
  • Label hierarchy
  • Label sentiment

Was this page helpful?

Get The Help You Need
Learning RPA - Automation Courses
UiPath Community Forum
Uipath Logo
Trust and Security
© 2005-2025 UiPath. All rights reserved.