# Reviewing label predictions

> :::note
You must have assigned the **Source - Read** and **Dataset - Review** permissions as an Automation Cloud user, or the **View sources** and **Review and annotate** permissions as a legacy user.
:::

:::note
You must have assigned the **Source - Read** and **Dataset - Review** permissions as an Automation Cloud user, or the **View sources** and **Review and annotate** permissions as a legacy user.
:::

## Annotating overview

After the **Discover** phase, the model will start making predictions for many of the labels in your taxonomy.

The purpose of the **Explore** phase is to review predictions for each label, confirming whether they are correct and correcting them where they are not. As a result, it provides many more training examples for the model.

Make sure you consider the following key actions when reviewing label predictions:

* Where the predictions are correct, you should confirm or accept them by selecting them.
* Where they are incorrect, you should either dismiss or ignore them, or, alternatively, add the correct labels that do apply. To add a different label, select the plus `+` button and enter the name of the label. This is the way to correct wrong predictions, by adding the correct one and not selecting on any incorrectly predicted labels.

The following images show how predictions appear in Communications Mining™ for data with and without sentiment. Hovering your mouse over the label will also show the confidence the model has that the specific label applies.

The transparency of the predicted label provides a visual indicator of the confidence of the model. The darker the colour, the higher the confidence, and vice versa:

  ![Message with predictions without sentiment enabled.](https://dev-assets.cms.uipath.com/assets/images/ixp/ixp-message-with-predictions-without-sentiment-enabled-322605-0607a4cc.webp)

  ![Message with predictions with sentiment enabled.](https://dev-assets.cms.uipath.com/assets/images/ixp/ixp-message-with-predictions-with-sentiment-enabled-322857-95b13a6e.webp)

To delete a label you accidentally applied, you can hover over it, and an X will appear. Select the X button to remove the label:

  ![The X button to remove a label that shows when hovering over a label.](https://dev-assets.cms.uipath.com/assets/images/ixp/ixp-the-x-button-to-remove-a-label-that-shows-when-hovering-over-a-label-322373-53dc77f8.webp)

:::important
Add all labels that apply to the message you are reviewing. Indicating the model that a label does not apply is just as important as indicating it what does.
:::