Demo Language

Workflow Path

Explanation Level

🔒

Why Deterministic Detection Matters

The scanner uses patterns to find personal information. Like a spell-checker that always finds "teh" and suggests "the"—it's consistent and reliable.

Same text + Same settings = Same results

No guessing. No "maybe". No surprises.

Regex-based detection ensures reproducible compliance audits. Unlike ML models that may vary results based on training data or random seeds, regex patterns produce identical outputs for identical inputs.

Benefits:

Audit trail integrity (results don't change between scans)
Regression testing (verify fixes don't break detection)
Legal defensibility (consistent application of detection rules)

The detection engine uses compiled regex patterns with zero ML/AI components. Pattern matching is deterministic by design:

// Example: German IBAN pattern
DE_IBAN: /DE\d{2}\s?\d{4}\s?\d{4}\s?\d{4}\s?\d{4}\s?\d{2}/g

// Input: "DE89 3704 0044 0532 0130 00"
// Output: Always matches at position 0, length 31
// No variance, no confidence scores, no probabilistic output

Hash verification: Same input text + same config.language → identical SHA-256 hash of detection results JSON.

Correct vs Wrong Language Comparison

✓ Correct Language

✗ Wrong Language

Why Deterministic Detection Matters