abishek-kronosx/hipaa-column-classifier

HIPAA Column Classifier

95%+ accuracy PII/PHI/O detector for database columns

Usage

from transformers import pipeline

classifier = pipeline(
    "text-classification",
    model="abishek-kronosx/hipaa-column-classifier"
)

print(classifier("memberCell"))     # [{'label': 'PII', 'score': 0.98}]
print(classifier("radiologyExam"))  # [{'label': 'PHI', 'score': 0.95}]
print(classifier("skuAddress"))     # [{'label': 'O', 'score': 0.99}]

| Column        | Prediction |
| ------------- | ---------- |
| memberCell    | PII ✅      |
| radiologyExam | PHI ✅      |
| drugCode      | PHI ✅      |
| skuAddress    | O ✅        |