README.md

metadata

license: apache-2.0
language:
  - en
  - ga
task_categories:
  - token-classification
tags:
  - pii
  - ireland
  - ppsn
  - eircode
  - synthetic
  - token-classification
pretty_name: OpenMed Irish PPSN Eircode Spec v1
configs:
  - config_name: default
    data_files:
      - split: train
        path: train.jsonl
      - split: validation
        path: validation.jsonl
      - split: test
        path: test.jsonl

OpenMed Irish PPSN Eircode Spec v1

Focused synthetic token-classification dataset for Irish PPSN and Eircode detection.

This repo contains synthetic training rows, not a fine-tuned model.

What A Row Looks Like

Each row uses a fixed schema:

• id: row id inside the split
• text: rendered text string
• tokens: tokenized text
• labels: BIO labels aligned to tokens
• language: en or ga
• source_dataset: generator identifier
• source_domain: optional domain tag, empty in this release
• source_type: optional subtype tag, empty in this release
• file_name: optional source file identifier, empty in this release
• sentence_number: optional source sentence index, -1 in this release

Coverage

This dataset focuses on:

• checksum-valid or format-valid PPSN positives
• spec-valid Eircode positives
• near-miss negatives around both formats
• English and Irish Gaelic contexts

Released labels:

B-PPSN, B-first_name, B-last_name, B-postcode, I-PPSN, I-last_name, I-postcode, O

Splits

• train: 12000
• validation: 1200
• test: 1200

Language Mix

Rows by language:

• en: 7212
• ga: 7188

Sources Used For Generation

• PPSN:
  • https://en.wikipedia.org/wiki/Personal_Public_Service_Number
  • https://www.revenue.ie/en/tax-professionals/tdm/income-tax-capital-gains-tax-corporation-tax/part-42/42-04-35a-20250214122127.pdf
• Eircode:
  • https://www.eircode.ie/what-is-eircode
  • https://www.eircode.ie/faqs
  • https://en.wikipedia.org/wiki/List_of_Eircode_routing_areas_in_Ireland

Intended Use

• bootstrap or harden Irish PPSN and Eircode token classifiers
• inspect format-driven synthetic coverage separately from broader mixed training sets
• reproduce the spec-driven component used in the public Irish CorePII training mix

Limitations

• synthetic data only
• narrow label scope; this is not a full Irish PII dataset
• public rules around PPSN suffix handling are imperfect, so the generator is conservative and operationally focused

License

This dataset is synthetic and released under Apache-2.0.