EU DSA Content Moderation Trends API

Turn the European Commission's official Digital Services Act Transparency Database aggregates into ranked, analysis-ready moderation trends.

The Actor downloads public monthly aggregate archives and groups millions of statements of reasons by platform, violation category, month or day, automation level, decision ground, content source, and moderation action.

What you can analyze

• Which platforms report the most moderation decisions
• The leading violation categories for a platform
• Fully automated versus partially or non-automated decisions
• Content removals, demotions, labels, account actions, and other measures
• Daily or monthly changes across up to 12 available months
• A platform, category, action, or decision-ground subset

Useful applications include trust-and-safety intelligence, policy research, regulatory monitoring, social-media analysis, journalism, and compliance dashboards.

Input

{
  "monthsBack": 3,
  "platforms": ["TikTok"],
  "categories": [],
  "automation": "all",
  "decisionGrounds": [],
  "actions": ["content_removed", "content_demoted", "content_labelled"],
  "groupBy": ["platform", "category", "month"],
  "maxResults": 100
}

monthTo is optional and uses YYYY-MM. When omitted, the Actor probes backward from the previous calendar month and skips archives that have not yet been published.

Platform and category filters use exact values from the official aggregate dataset. Leave them empty to include all values.

Output

Each dataset item is one ranked aggregate row:

{
  "platform": "TikTok",
  "category": "STATEMENT_CATEGORY_SCAMS_AND_FRAUD",
  "date": null,
  "month": "2026-05",
  "automation": null,
  "decisionGround": null,
  "sourceType": null,
  "action": null,
  "statementCount": 426322,
  "shareOfFilteredStatements": 0.01130007,
  "monthFrom": "2026-05",
  "monthTo": "2026-05",
  "sourceUrl": "https://transparency.dsa.ec.europa.eu/explore-data/download",
  "retrievedAt": "2026-07-23T17:00:00.000Z"
}

Results are sorted by statementCount. shareOfFilteredStatements is the row's fraction of all statements that passed the filters. When grouping by action, one underlying statement can contain multiple actions, so action shares may overlap and need not sum to 1.

The REPORT record in the default key-value store includes available and missing months, source archive URLs, raw and matched aggregate row counts, returned and charged results, filters, and completion status.

API example

curl "https://api.apify.com/v2/acts/YOUR_USERNAME~eu-dsa-content-moderation-trends-api/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"monthsBack":1,"monthTo":"2026-05","platforms":["TikTok"],"groupBy":["platform","category","month"],"maxResults":20}'

Reliability and safety

• Uses public monthly aggregates from the official European Commission DSA Transparency Database
• Enforces strict compressed and uncompressed archive limits
• Rejects unsafe ZIP paths and malformed CSV structures
• Handles the source's escaped platform names
• Limits each run to 12 months and 500 output rows
• Emits only aggregate statistics, not user-level content
• Writes data only after filters and ranking have completed

The Actor is an independent analytics tool and is not affiliated with or endorsed by the European Commission. It does not provide legal advice. Always follow the official source link when a result is used for compliance or research decisions.

Official source

• DSA Transparency Database data download: https://transparency.dsa.ec.europa.eu/explore-data/download
• Official data documentation: https://dsa.pages.code.europa.eu/transparency-database/dsa-tdb/data_sources.html