[QUERY] Canadian Personal Health Numbers not being detected by Text Analytics RecognizePiiEntitiesAsync method

[YEGCSharpDev]

Library name and version

Azure.AI.TextAnlaytics

Query/Question

When sending Canadian PHN(Personal Health Number)s, both in province specific format and without a defined format, PHNs are not detected by RecognizePiiEntitiesAsync method.

SIN (Social Insurance Number) is being detected with 96% confidence score. Sometimes PHN is also being picked up as SIN, but 100% of the time no PHN is being detected.

Formats for PHN from all provinces were fed to the client and none of them were detected.

The following categories were added when calling RecognizePiiEntitiesAsync method.

• PiiEntityCategory.CASocialInsuranceNumber
• PiiEntityCategory.CAPersonalHealthIdentification
• PiiEntityCategory.CAHealthServiceNumber

We are wondering if the service is being used correctly by us. Here is the snippet of code that calls the service and returns results.

RecognizePiiEntitiesOptions options = new() {CategoriesFilter = { PiiEntityCategory.CAHealthServiceNumber ,PiiEntityCategory.CASocialInsuranceNumber, PiiEntityCategory.CAPersonalHealthIdentification } };
PiiEntityCollection entities = await client.RecognizePiiEntitiesAsync(document: documentData, options: options);
    if (entities.Count > 0)
        {
            foreach (PiiEntity entity in entities)
            {
                textsToRedact.Add(entity.Text);
                Console.WriteLine($"Text: {entity.Text}, Category: {entity.Category}, SubCategory: {entity.SubCategory}, Confidence score: {entity.ConfidenceScore}");
            }
        }

Sample PHNs sent to the service (none of these values are actual values)

• 893 615 460
• 5584-486-674-YM
• BOUF 9010 1516
• 398 516 364
• 465 978 309 317
• N6739906
• 5004 739 743
• 957464812

Environment

OS and .NET runtime version

OS :

.NET SDK: Version: 8.0.101 Commit: 6eceda187b Workload version: 8.0.100-manifests.30fce108

Runtime Environment:

OS Name: Windows OS Version: 10.0.19045 OS Platform: Windows RID: win-x64 Base Path: C:\Program Files\dotnet\sdk\8.0.101\

Host:

Version: 8.0.1 Architecture: x64 Commit: bf5e279d92 .NET SDKs installed: 8.0.101 [C:\Program Files\dotnet\sdk]

IDE

Visual Studio V17.8.6

[github-actions[bot]]

Thanks for the feedback! We are routing this to the appropriate team for follow-up. cc @assafi @quentinRobinson @wangyuantao.

[YEGCSharpDev]

It's been a few months since this issue was opened. Following up.