Australian Patterns

Allyanonimiser includes specialized patterns for detecting Australian-specific PII and identifiers.

Australian Tax File Number (AU_TFN)

Tax File Numbers are unique identifiers issued by the Australian Taxation Office.

Pattern Example: - 123 456 782 (checksum-valid — use this shape in tests) - TFN: 123 456 782

Regex Pattern:

\b\d{3}\s*\d{3}\s*\d{3}\b

Contexts: "tax", "file", "number", "tfn"

Checksum validation (since v3.2.0): detected TFNs are re-checked against the ATO's modulus-11 weighted algorithm. Numbers that match the regex but fail the checksum — like 123 456 789 — are filtered out. Use 123 456 782 in your own test data.

Australian Phone Number (AU_PHONE)

Australian mobile and landline phone numbers.

Pattern Example: - 0412 345 678 - +61 412 345 678 - (02) 9876 5432

Regex Pattern:

\b(?:\+?61|0)4\d{2}\s*\d{3}\s*\d{3}\b

Contexts: "phone", "mobile", "call", "contact"

Australian Medicare Number (AU_MEDICARE)

Medicare numbers issued by Services Australia. The first digit must be 2–6 and the last digit (the individual reference number, IRN) must be 1–9.

Pattern Example: - 2123 45678 1 (starts with 2, IRN is 1) - 5123 45678 1 - Medicare: 2345 67890 1

Regex Pattern:

\b[2-6]\d{3}\s*\d{5}\s*[1-9]\b

Contexts: "medicare", "card", "health", "insurance"

Validation: detected candidates are re-checked against the format rules (first digit 2–6, IRN 1–9). Numbers like 1234 56789 1 (starts with 1) or 2123 45678 0 (IRN is 0) are filtered out.

Australian Driver's License (AU_DRIVERS_LICENSE)

Driver's license numbers from various Australian states.

Pattern Example: - 12345678 (NSW) - 1234AB (NSW legacy)

Regex Patterns:

\b\d{8}\b
\b\d{4}[a-zA-Z]{2}\b

Contexts: "license", "licence", "driver", "driving"

Australian Address (AU_ADDRESS)

Australian address formats including street addresses and postal codes.

Pattern Example: - 42 Main St, Sydney NSW 2000 - Unit 5, 123 Example Road, Melbourne VIC 3000

Regex Patterns: Multiple complex patterns for detecting address components.

Contexts: "street", "road", "avenue", "address", "suburb"

Australian Postcode (AU_POSTCODE)

Four-digit Australian postcodes.

Pattern Example: - 2000 - 3000 - NSW 2000

Regex Pattern:

\b\d{4}\b

Contexts: "postcode", "post", "code", "postal"

Australian BSB and Account Number (AU_BSB_ACCOUNT)

BSB (Bank-State-Branch) and account number combinations.

Pattern Example: - 123-456 12345678 - 123456 12345678

Regex Pattern:

\b\d{3}-\d{3}\s*\d{6,10}\b

Contexts: "bsb", "account", "banking", "bank"

Australian Business Number (AU_ABN)

Australian Business Numbers issued by the Australian Business Register.

Pattern Example: - 51 824 753 556 (the ATO's published test ABN — checksum-valid) - ABN: 51 824 753 556

Regex Pattern:

\b\d{2}\s*\d{3}\s*\d{3}\s*\d{3}\b

Contexts: "abn", "business", "number", "australian"

Australian Passport Number (AU_PASSPORT)

Australian passport numbers.

Pattern Example: - N1234567 - PA1234567

Regex Pattern:

\b[A-Za-z]\d{8}\b

Contexts: "passport", "travel", "document", "international"

Usage Example

from allyanonimiser import create_allyanonimiser

# Create the analyzer
ally = create_allyanonimiser()

# Example text with Australian PII
text = """
The customer John Smith with TFN 123 456 782 and Medicare 1234 56789 1 
called from 0412 345 678 regarding their policy. They live at 
42 Main St, Sydney NSW 2000 and their ABN is 12 345 678 901.
"""

# Analyze the text
results = ally.analyze(text)

# Filter for Australian patterns only
au_results = [r for r in results if r.entity_type.startswith("AU_")]

# Print the results
for result in au_results:
    print(f"Entity: {result.entity_type}, Text: {result.text}, Score: {result.score}")