PII Data Deletion

If your submissions contain personally identifiable information (PII), you may want to delete it from submissions for security reasons, or your organization may have regulations that require you to do so. In any case, Hyperscience can automatically delete PII from your submissions at the time you specify. If you enable PII data deletion, all document image data, including original uploaded images and any processed or corrected images, will be deleted.

Settings

You can manage PII data deletion in your instance by configuring the following settings, which are available in  the “General” section of the application settings (Administration > System Settings). 

PII Data deletion (images and fields only) setting

The PII Data deletion (Images and fields only) setting  is disabled by default. When you enable it, you will also need to set your PII deletion policy. 

PII Deletion policy

You can choose to delete PII from submissions a certain number of days after either their completion dates (Submission Complete Date) or their submitted dates (Date Submitted). You can also specify the time of day that the system will delete PII.

If you choose Date Submitted as the basis for your PII deletion window, you should make your window long enough to ensure that you are only deleting PII from processed submissions. Otherwise, outstanding manual tasks or processing submissions may be deleted. 

Note that if you choose Date Submitted as the basis for your PII deletion window, the PII deletion policy will also delete all training documents that are uploaded through the Classification tab of the Model Details page. If you want to exclude these training documents from the PII deletion policy, you need to:

  1. Go to /admin/common/freeformconfig/.

  2. Click the Edit config link.

  3. Enable the Retain Pii For Nlc Training Data setting.

  4. Click Save Changes.

Any changes you make to your PII deletion policy will be applied retroactively to all qualifying submissions in the systems. PII may not be deleted immediately after saving your settings, as PII deletion takes place once each day at the time you specify.

Submission record deletion setting

If you are deleting PII data, you can also choose to delete submissions that have had their PII removed and have already been used in Transcription Automation Training, if enabled.

To automatically delete these submissions, enable the Submission record deletion setting. When you enable this setting, you will also need to set your submission record deletion policy. 

Submission record deletion policy

You can specify how long after PII data deletion or Transcription Automation Training to delete submissions. The system will add the number of days you enter to either the PII data deletion window or the Transcription Automation Training window, whichever is greater. You have separate Transcription Automation Training windows for Structured and Semi-structured document transcriptions. A Transcription Automation Training window is the value in the Periods of records to use flow settings, which can be found in the “Structured Document Transcription” and “Semi-structured Document Transcription” sections. Your Transcription Automation Training window is the value in the Period of records to use setting, which is in the “General Transcription” section of your flow's settings.

Examples

Let's say your PII data deletion window is 60 days after the submission completion date and your Transcription Automation Training window for Structured documents is 30 days. If you enter a window of 10 days in your submission record deletion policy, the system will delete submissions 70 days after the Submission completion date (60 + 10 days).

In another case, say your PII data deletion window is 30 days after the submission completion date and your Transcription Automation Training window for Semi-structured documents is 50 days. If you enter a window of 10 days in your submission record deletion policy, the system will delete submissions 60 days after the Submission completion date (50 + 10 days).

You can also choose to delete submissions at a certain time of day and for the duration you specify.

Any changes you make to your submission record deletion settings will be applied retroactively to all qualifying submissions in the system. Submissions may not be deleted immediately after saving your settings, as submission deletion takes place once each day at the time you specify.

Implications for upgrades

If you are planning to upgrade to a new version of Hyperscience, you should consider how your PII deletion policy may impact automation after upgrading. To learn more, see Upgrade Best Practices

Implications for reporting

When you delete PII from submissions and retain the submissions after PII deletion, those submissions will still be included in the calculation of reporting statistics. However, if you enable Submission record deletion, the deleted submissions will not be included in reporting statistics after their deletion. Therefore, you may notice a drop in some reporting metrics after enabling Submission record deletion. Specifically, submission deletion will affect the following reports:

  • Reporting > Accuracy

    • System Transcription Sampled Errors

    • Field Exception Report

  • Reporting > Processing Time

    • Submissions SLA

  • Reporting > User Performance

    • Transcription Sampled Errors