# Recommended Mode

In the recommended mode of Cybral Guard, users play an essential role in file classification while benefiting from AI-powered assistance. In this mode, users are tasked with manually assigning labels to files, ensuring accurate classification based on their expertise and decision. However, should a user manage labeling a file, the AI system steps in to analyze its content and automatically apply an appropriate label. By easily blending user input with AI-driven automation, the recommended mode guarantees a full file classification, safeguarding data integrity and compliance with organizational standards.

<figure><img src="/files/TtXAyNGp4HOP61NsHJe0" alt="" width="244"><figcaption><p>Recommended Mode Classification Group</p></figcaption></figure>

In the recommended classification mode, the AI's intervention is mandatory. This mode cannot be changed, as it ensures that the AI system automatically assigns labels to files when users forget to do so manually.

<figure><img src="/files/Ccx0P3TG62b8p35Gtdt7" alt="" width="375"><figcaption><p>Classification Mandatory</p></figcaption></figure>

The label prioritization feature is available across all classification modes. This option can be enabled by checking the corresponding box provided. When activated, users can customize the priority order of labels within each classification group

<figure><img src="/files/SsSEkTA1rZFR4joZZNUF" alt="" width="194"><figcaption><p>Label prioritization</p></figcaption></figure>

When an administrator selects the recommended classification mode, they gain access to the AI classification page, where the classification group will be visible. This group will be listed under both manual and AI classification categories, reflecting its usage in the recommended mode.

<figure><img src="/files/a9eYa0IMmjvnL9HmlsgC" alt="" width="375"><figcaption><p>AI Classification</p></figcaption></figure>

To initiate AI training for automatic file labeling, the administrator should navigate to the "Manage Training Data" section. This feature allows administrators to upload sample files that the AI will use to learn and generate accurate labels automatically.

<figure><img src="/files/Zw9rh2d7a1SHs6YJ3Amr" alt="" width="301"><figcaption><p>Train New Classifier</p></figcaption></figure>

By accessing the "Manage Training Data" feature, administrators can upload the necessary files for AI training on each label. This step ensures that the AI system is equipped with sufficient data to learn and accurately classify files based on predefined labels.

<figure><img src="/files/OrcvpngF40qHayBSz8th" alt="" width="563"><figcaption><p>Manage Training Data</p></figcaption></figure>

When administrators choose the recommended classification mode, they will need to follow the same [training](/business/classification-groups/ai-mode.md) process as in the AI classification mode.

<figure><img src="/files/hPOwXfLYWQT8G4NrIDT8" alt="" width="375"><figcaption><p>Training Data Management</p></figcaption></figure>

In the recommended classification mode, particularly in the manual aspect of it, users will also adhere to the AI data classification discussed in the Manual mode ([AI Data Classification](/business/classification-groups/manual-mode.md)). This means that while users have the responsibility to manually label files, they will also rely on the AI's insights and recommendations for accurate classification.

{% hint style="success" %}
In recommended mode, users are required to manually add labels to files. However, if a user forgets to label a file, the AI system intervenes and automatically assigns a label based on its analysis of the file content.
{% endhint %}


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://manual.cybral.com/business/classification-groups/recommended-mode.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.