Pii Masking Digital Pdi 350K: Global Personal Information for PII Masking

A global corpus of 369,310 text entries containing 1,281,920 annotations for 28 types of Personally Identifiable Information (PII). The dataset, part of the PII-Masking-3M collection by Ai4Privacy, covers 30 languages and 37 regions. It was last updated on June 3, 2026.

Use Cases

• Train PII detection models based on the 28 defined PII labels.
• Benchmark multilingual NER systems based on text in 30 languages.
• Develop data masking or anonymization tools based on the annotated PII spans.
• Study regional variations in PII formats based on coverage of 37 regions.

Strengths

• Large annotation volume with over 1.28 million PII annotations.
• Broad multilingual coverage across 30 languages.
• Wide regional scope covering 37 regions.
• Part of a larger, defined corpus (PII-Masking-3M) with a clear taxonomy.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Ai4Privacy (ai4privacy)

Freshness

Last updated 2026-06-03 07:27:17; freshness should be verified.

Geography

Global, with specific mention of Asia Pacific regions.