European Personal Information Redaction Sample

A preview of 50 sample entries from the PII-Masking-2M dataset for European languages, created by AI4Privacy. The dataset contains text with redacted personal identifiable information. The full dataset was referenced in 2026.

Use Cases

• Training models to identify PII spans using the provided redacted text and label distribution.
• Benchmarking multilingual PII detection performance across the specified European language distribution.
• Analyzing the geographic coverage of PII types within the European coverage data.

Strengths

• Part of a larger 2-million entry corpus (PII-Masking-2M).
• Provides explicit label and language distribution statistics.

Limitations

• Contains only 50 sample entries, insufficient for model training.
• Source text and actual PII values are redacted, preventing direct analysis of PII patterns.
• Specific column names and data structure are unknown.

Provenance

Source

AI4Privacy.

Collection Method

null

Time Range

null

Freshness

Last updated in April 2026.

Geography

European coverage.