anonym.legal

By · Last updated 2026-06-05

Terug na BlogGDPR & Nakoming

NAIH Hongarye: TAJ-Szam en Adoazonosito Jel

Hongaarse NER-akkuraatheid is 67% teenoor die EU-gemiddeld van 82% - NAIH se 2024-beoordeling. TAJ-szam geweegde kontrolesom en adoazonosito jel-opsporingsgapings.

June 5, 20267 min lees
Hungary NAIHTAJ-szám detectionHungarian NERHungarian GDPR complianceAI DPIA

NAIH Hongarye: TAJ-Szam en GDPR-tegniese Vereistes

Opgedateer vir 2026

Hongarye se dataowerheid is NAIH. Sy 2024-verslag het gevind dat NER-akkuraatheid vir Hongaars slegs 67% is. Die EU-gemiddeld is 82%. Hierdie gaping skep werklike risiko. Instrumente wat vir Engels of Duits gebou is, mis Hongaarse identifiseerders teen hoe koerse.

Hoekom Hongaarse NER Laag Aangeteken Word

Drie kenmerke van Hongaars breek standaard NLP-modelle.

Agglutinasie: Hongaars voeg agtervoegsels by wortelwoorde. Dieselfde naam neem baie vorms in 'n sin aan. "Kovacs Peter" in subjekposisie word "Kovacs Peternek" in 'n ander rol. NER-modelle moet al hierdie vorms aan een persoon koppel.

Naamvolgorde: Hongaars stel die familienaam eerste. Die meeste NLP-modelle verwag eerste naam eerste. Hierdie omgekeerde volgorde veroorsaak gemiste opsporings.

Spesiale karakters: Hongaars gebruik o met dubbelaksent en u met dubbelaksent. Dit is nie dieselfde as Duitse umlaut nie. Gemengde kodering - Windows-1250 teenoor UTF-8 - veroorsaak ook mislukkings.

Hierdie drie faktore verklaar die meeste van die akkuraatheidsgaping in NAIH se 2024-verslag.

TAJ-Szam: Hongarye se Sosiale Sekerheids-nommer

Die TAJ-szam (Tarsadalombiztositasi Azonosito Jel) is 'n 9-syfer-nommer. Dit verskyn in gesondheidsorg-, salaris-, maatskaplike voordele- en pensioenrekords.

Kontrolesom: Vermenigvuldig syfers 1 tot 8 met gewigte 3, 7, 3, 7, 3, 7, 3, 7. Tel die resultate op. Neem modulo 10. Dit gee die toetssyfer.

Hierdie algoritme is uniek aan Hongarye. Dit is nie dieselfde as die Luhn-algoritme wat in ander lande gebruik word nie.

Generiese instrumente spoor TAJ-szam met slegs 61% akkuraatheid op, volgens die NAIH 2024-verslag. Die 9-syfer-formaat lyk soos baie ander nommers in Hongaarse dokumente. Sonder die kontrolesom-stap merk instrumente vals positiewe en mis werklike nommers.

Adoazonosito Jel: Hongarye se Belasting-ID

Die adoazonosito jel is 'n 10-syfer persoonlike belastingnommer. Die eerste syfer is altyd 8. Dit verskyn in indiensnemingsrekords, belastingliasserings en finansiele dokumente.

Kontrolesom: Neem syfers 2 tot 9. Vermenigvuldig met gewigte 9, 7, 3, 1, 9, 7, 3, 1. Tel die resultate op. Neem modulo 10. Dit is die toetssyfer. 'n Resultaat van 0 beteken die toetssyfer is 0.

NAIH-handhawingsgevalle toon dat hierdie nommer dikwels in HR-dokumente gemis word wanneer instrumente vir ander tale opgestel is.

Sien ons EU nasionale belasting-ID-gids vir hoe hierdie nommers oor lidstate vergelyk.

NAIH se DPIA-vereiste vir KI-stelsels

NAIH se 2024-leiding vereis 'n voltooide DPIA voor enige KI-stelsel persoonlike data verwerk. Dit is strenger as die algemene GDPR-toets. Die DPIA moet dek:

  1. Datavloei - opleidingsdata, insette en uitsette
  2. Regsbasis - gedokumenteer vir elke aktiwiteit
  3. Taalakkuraatheid - vereis vir tale onder die EU-gemiddeld
  4. Menslike hersiening - 'n manier om geoutomatiseerde besluite te kontroleer

Die DPIA moet elke jaar opgedateer word wanneer die stelsel opgelei word.

Vir spanne wat KI-instrumente op Hongaarse data ontplooi, is die volgorde vas: DPIA eerste, dan ontplooiing.

Minimum Tegniese Beheermaatreels

Drie beheermaatreels vorm die basislyn vir NAIH-nakoming:

  1. TAJ-szam-opsporing met modulo-10-kontrolesom - patroonmassering alleen is nie genoeg nie
  2. Adoazonosito jel-opsporing met kontrolesom-validasie - krities vir HR en finansies
  3. Hongaarse NER met agglutinasie-ondersteuning - moet o met dubbelaksent, u met dubbelaksent en koderingsvariante hanteer

Sien ons BFDI Duitsland-gids om te vergelyk hoe Sentraal-Europese DPA's tegniese vereistes stel. Vir 'n soortgelyke taalkloof in Sentraal-Europa, sien ons Tsjeggiese UOOU-gids.

Bronne

Verwante Artikels

GDPR & Nakoming

Japan My Number: Verhoeff & APPI

63% of generic tools fail My Number detection in Japanese documents. My Number uses Verhoeff algorithm — the most complex national ID checksum in Asia.

GDPR & Nakoming

HDPA Greece: AFM & AMKA Detection

Greek AFM detected with 52% accuracy by generic tools. HDPA issued 89 decisions in 2024 — up 162% from 2022. Tourism and maritime sectors face distinct.

GDPR & Nakoming

Czech Rodné Číslo: Gender Encoding and GDPR

Czech rodné číslo encodes gender via 50-offset month encoding — making it GDPR Article 9 special category data. 67% of Czech firms use German tools.

Gereed om u data te beskerm?

Begin om PII te anonimiseer met 285+ entiteitstipes in 48 tale.

Begin Gratis ProeflopieBesoek Kenmerke

About this page

We update this page when our platform or the law changes.

Read our founder note for how we work.

Each change shows up in the timestamp at the top.

Related reading

We follow these rules

  • GDPR (EU 2016/679).
  • ISO/IEC 27001:2022.
  • NIS2 (EU 2022/2555).
  • HIPAA safe harbor under 45 CFR § 164.514(b)(2).

Our promise

We do not sell your data.

We do not train models on your text.

We store your files in Germany.

You can delete your account at any time.

You own your work.

Where we run

Our servers live in Falkenstein, Germany.

We use Hetzner. They hold ISO 27001 certification.

All data stays in the EU.

Backups run every day.

Need help?

Email support@anonym.legal.

We reply within one business day.

How we test

We run a full check suite on every release.

Each surface gets its own sweep script and report.

Human reviewers spot-check the output each week.

We track recall and precision on a labelled set.

Bad runs block the deploy.

What we never do

  • We never sell your information to third parties.
  • We never train models on what you upload.
  • We never keep your work after you delete it.
  • We never share keys with any outside firm.
  • We never run ads inside the product.

Plans in plain words

We sell credits, not seats.

One credit covers one short job.

Long jobs use a few credits each.

You can top up at any time.

Unused credits roll over each month.

Read the plans page for current rates.

Who built this

A small team of engineers and lawyers built this.

We ship from Europe and work in the open.

Our founder note spells out why we started.

Where to start

How the parts fit

A browser add-on cleans text inside Chrome.

A Word plug-in handles drafts in Office.

A small desktop tool works on whole folders.

An agent protocol link feeds large models safely.

All four share one core engine and one rule set.

Words from our team

We started this work after a lunch about cookies.

One friend kept getting odd ads on her phone.

We asked why a court file leaked through a draft.

We sketched the first build on a napkin that week.

By month three we had a tiny demo for a friend.

She used it on her first case the next day.

Common questions we hear

Can the tool read scanned PDFs? Yes, with OCR.

Does it work on long files? Yes, in small chunks.

Can I roll my own rule set? Yes, save it as a preset.

Does it run offline? The desktop build runs offline.

Do you keep my files? No, the cloud build wipes after each run.

Will it learn from my work? No, we never train on inputs.

A short tour of the workflow

Upload a file or paste a snippet of prose.

Pick the entities you want gone from the draft.

Choose a method: replace, mask, hash, encrypt, or redact.

Press run and watch the side panel show each hit.

Skim the result and tweak any rule that misfired.

Save the cleaned file or send it to a teammate.