Personal Work Information Masking Preview for Text Anonymization

EPII Personal Work Information (PWI) Masking Preview Dataset contains 400 text samples for identifying and masking sensitive personal work information. The preview demonstrates the structure and label taxonomy of a commercial dataset containing over 100,000 entries, created by ai4privacy. The dataset page was last updated in April 2026.

Use Cases

• Train a named entity recognition model to detect PWI labels like 'job_title' or 'employer' within text snippets.
• Fine-tune a transformer model for the text masking task using the provided annotation taxonomy.
• Benchmark the performance of privacy-preserving text processing algorithms against the annotated PWI entities.
• Develop a sequence labeling model to identify the boundaries of sensitive work information spans in documents.

Strengths

• Preview contains 400 samples for initial evaluation.
• Represents the structure and taxonomy of a larger dataset with over 100,000 entries.

Limitations

• The accessible dataset is only a 400-sample preview, not the full commercial collection.
• Specific column names, data formats, and the exact row count of the preview are not provided.

Provenance

Source

ai4privacy

Collection Method

null

Time Range

null

Freshness

null

Geography

null