> For the complete documentation index, see [llms.txt](https://docs.collibra.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.collibra.com/data-quality-self-hosted/core-concepts/findings/data-quality-monitors/record.md).

# Record

The Record monitor detects rows of data that drop out of or are added to a dataset and down-score any observations. To see record findings, select at least one numeric data type column in the [Outliers layer](/data-quality-self-hosted/core-concepts/explorer/create-a-dq-job/add-layers.md#tab-3), a Key column, and a Date column. If you do not select at least one numeric column, a Key column, and a Date column, Collibra DQ skips the record check activity. Alternatively, you can exclude Outlier findings and detect only Records by selecting a Key column and a Date column but excluding the numeric data type column from your Outlier configuration.

The following table describes the information available for the Record tab on the Findings page.

<table><thead><tr><th width="152">Column</th><th>Description</th></tr></thead><tbody><tr><td>Observation</td><td>The type of record anomaly detected.</td></tr><tr><td>Status</td><td>Allows you to <a href="/data-quality-self-hosted/core-concepts/findings/working-with-assignments/validating-findings.md">validate</a> or <a href="/data-quality-self-hosted/core-concepts/findings/working-with-assignments/resolving-findings.md">resolve</a> an observation and, when applicable, assign it to a user for further analysis.</td></tr><tr><td>Profile</td><td><p>The user account that is assigned to this record finding. When the Status is Assigned, a user profile displays in this column.</p><div data-gb-custom-block data-tag="hint" data-style="info" class="hint hint-info"><p>When a record finding is unassigned, the profile column is empty.</p></div></td></tr></tbody></table>

## Retrain

Updates the DQ models used for automated data-quality checks. The Retrain process consumes current training data (and optionally labels/config), builds a new model or rule-set, validates it, and publishes a new version that will be used to score incoming data and compute DQ metrics. Use **Retrain** when data distributions change, new labels/ground truth are available, business rules change, or on a scheduled cadence to prevent model drift.