anonym.legal

By · Last updated 2026-02-26

Terug naar BlogTechnisch

Meertalige NER: Arabisch en Chinees

Arabische NER bereikt F1-scores van 0,60-0,83 in ACL 2024-benchmarks. Chinese segmentatie vereist woordgrensdetectie die de meeste Europese NER-modellen missen.

February 26, 20268 min lezen
NERmultilingualArabic NLPChinese NLPPII detection

De Taalkloof in Named Entity Recognition

Named Entity Recognition (NER) — het detecteren van namen, organisaties, locaties en andere entiteiten in tekst — werkt aanzienlijk slechter voor niet-Engelse talen, met name voor talen met andere schriftsystemen.

ACL 2024-benchmarks vonden F1-scores van 0,60-0,83 voor Arabisch en 0,65-0,80 voor Chinees — aanzienlijk lager dan Engels (0,85-0,92) voor vergelijkbare documenttypes.

Die kloof heeft directe nalevingsimplicaties. GDPR beschermt persoonsgegevens ongeacht de taal. Als uw NER-tool Arabische of Chinese PII mist, heeft u geen GDPR-naleving bereikt voor die documenten.

Arabische NER Uitdagingen

Schriftrichting: Arabisch wordt van rechts naar links geschreven. Veel NLP-pipelines zijn gebouwd voor links-naar-rechts tekst.

Morfologische rijkheid: Arabisch combineert prefixes, suffixes en wortelwijzigingen op manieren die Europese talen niet doen.

Dialectvariatie: Modern Standaardarabisch, Egyptisch Arabisch en andere dialecten variëren aanzienlijk.

Ontbrekende diacritics: Informele Arabische tekst laat vaak korte klinkers weg, waardoor naamherkenning contextafhankelijk wordt.

Chinese NER Uitdagingen

Geen woordgrenzen: Chinees schrift heeft geen spaties tussen woorden. NER moet eerst woordsegmentatie uitvoeren.

Naamformaat: Chinese namen volgen patronen die anders zijn dan westerse namensconventies.

Traditioneel vs. Vereenvoudigd: Tools moeten beide schriftsystemen verwerken.

Gemengde tekst: Zakelijke documenten combineren vaak Chinese tekst met Engelse termen en technische terminologie.

Implicaties voor GDPR-naleving

Organisaties die Arabisch- of Chinestalige documenten verwerken, moeten expliciete bevestiging vragen van hun PII-leverancier over taalondersteuning en nauwkeurigheidsniveaus.

Bekijk de entiteitsdetectiepagina voor een overzicht van ondersteunde talen.

Bronnen

Gerelateerde Artikelen

Technisch

Cross-Platform PII: Mac, Linux, and Windows

Privacy officers on Mac, legal on Windows, data engineers on Linux — all processing the same data with different tools. Here's why OS-agnostic detection.

Technisch

Cross-Application PII: Word, Chrome, and AI

Customer data flows from browser research to Word drafts to Claude prompts. Each context switch is a potential leakage point.

Technisch

GDPR in App Logs: JSON PII Compliance

Application logs contain customer email addresses, IPs, and account numbers that GDPR Article 5(1)(e) requires be managed.

Klaar om uw gegevens te beschermen?

Begin met het anonimiseren van PII met 285+ entiteitstypen in 48 talen.

Start Gratis ProefperiodeBekijk Kenmerken

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.