Wikipedia Entity Intelligence — Facts & Trends (AI)

Hand it a name, get back an entity. Give this Actor a list of companies, people, products or topics and it resolves each one to its Wikipedia article, pulls a clean summary, extracts structured facts from Wikidata (founded, industry, headquarters, key people, employees, website, country), measures a 60-day pageview trend, and uses OpenAI to write a short, grounded brief — the category, the key facts, and what the attention curve is saying.

You bring the names. It hands back a knowledge record for each.

---

The flow

┌──────────────► Wikipedia summary (extract · description · Q-id)
  "Stripe" ─┤
  "Elon Musk"─────────────► Wikidata facts (founded · industry · HQ · CEO · …)
  "Kubernetes"             
            └──────────────► Wikimedia pageviews (60-day total + trend)
                                         │
                                         ▼
                                   AI brief (OpenAI)
                               category · keyFacts · attentionSignal

1. Resolve — each name is searched against Wikipedia to find the best-matching article title (no-match is handled gracefully).
2. Summarize — the REST summary endpoint gives a trimmed extract, a one-line description, the page type, the canonical URL and the Wikidata Q-id.
3. Fact-extract — when there's a Q-id, Wikidata claims are parsed for the common properties and any referenced entities are resolved to human labels in one batched call.
4. Trend — the last 60 days of daily pageviews are summed and split into first-30 vs last-30 to classify the trend as rising / falling / stable.
5. Brief — OpenAI (gpt-4o-mini, fast and low-cost) writes a 2–3 sentence brief, a category, key facts and an attention signal — grounded only in the fetched data. No invented facts.

---

What you get per entity

For every entity the Actor returns:

1. resolvedTitle and pageUrl — the exact Wikipedia article it matched, plus the canonical desktop link.
2. wikidataId, description and extract — the Q-id, the short description line and the trimmed summary paragraph.
3. Structured facts from Wikidata — founded, industry, headquarters, foundedBy, ceo, employees, website, country, instanceOf. Anything not stated stays null.
4. pageviews60d and pageviewTrend — total Wikipedia pageviews over the last 60 days and whether attention is rising, falling or stable.
5. AI layer — category (person / company / product / technology / place / other), keyFacts (short strings built only from the data), attentionSignal (one sentence on the trend), and aiBrief (2–3 sentences).

If a name has no Wikipedia match, you still get a record with error set so nothing silently disappears.

---

Three ways people use it

• Entity enrichment — turn a raw list of company or product names into structured entity data: founding year, industry, HQ and key people pulled straight from Wikidata, ready to merge into your CRM or data warehouse. A free, transparent alternative for firmographic-style enrichment.
• Knowledge-graph building — resolve names to stable Wikidata Q-ids and the canonical Wikipedia API record, then wire the instanceOf / industry / country relations into your knowledge graph without hand-mapping entities.
• Research & brand monitoring — track the 60-day pageviews trend for the people, products and companies you care about, with an AI-written brief that tells you at a glance what each one is and whether public attention is climbing.

Keywords this Actor is built around: Wikipedia API, Wikidata, entity data, pageviews, knowledge graph.

---

Input

Field	Type	Notes
entities	array	Required. Names to look up — companies, people, products, topics. One per line.
lang	string	Wikipedia language edition code. Default en.
includePageviews	boolean	Fetch the 60-day pageview trend. Default true.
maxResults	integer	Cap how many entities to process. Default 200.
maxConcurrency	integer	Entities processed in parallel. Default 4.
model	string	Override the OpenAI model. Default gpt-4o-mini.

Minimal input example

{
  "entities": ["Stripe", "Elon Musk", "Kubernetes"]
}

---

Output example

{
  "input": "Stripe",
  "resolvedTitle": "Stripe, Inc.",
  "wikidataId": "Q2944583",
  "description": "American financial services company",
  "extract": "Stripe, Inc. is an Irish-American multinational financial services and software as a service company ...",
  "category": "company",
  "founded": "2010",
  "industry": "financial technology",
  "headquarters": "South San Francisco",
  "foundedBy": "Patrick Collison",
  "ceo": "Patrick Collison",
  "employees": "8000",
  "website": "https://stripe.com",
  "country": "United States of America",
  "instanceOf": "business",
  "pageUrl": "https://en.wikipedia.org/wiki/Stripe,_Inc.",
  "pageviews60d": 184230,
  "pageviewTrend": "rising",
  "keyFacts": [
    "Founded in 2010 by Patrick and John Collison",
    "Operates in financial technology / payments",
    "Headquartered in South San Francisco"
  ],
  "attentionSignal": "Pageviews rose over the last 60 days, suggesting increased public interest.",
  "aiBrief": "Stripe is an American fintech company founded in 2010 that provides payment-processing software and APIs for online businesses. It is one of the most valuable private technology companies, serving merchants worldwide."
}

---

Pay-per-event pricing

This Actor uses pay-per-event billing with a single event:

Event	Charged when
entity-analyzed	Once per entity resolved across Wikipedia and Wikidata with facts, pageview trend and an AI brief. $0.10

The Actor logs its OpenAI token usage and the estimated AI cost at the end of every run, so the event price always stays safely above the model cost.

---

Data source

All data comes from the public, free Wikipedia, Wikidata and Wikimedia REST/Action APIs — no API key for the data itself. As the Wikimedia APIs require, every request is sent with a descriptive User-Agent identifying this Actor, so usage stays within their policy. Facts reflect whatever the community has published; missing claims come back as null rather than being guessed.

---

Setup

The AI brief calls the OpenAI API. Set the environment variable / secret OPENAI_API_KEY on the Actor. The Wikipedia, Wikidata and pageview data require no key.

---

Use it from an AI agent (Apify MCP server)

This Actor is available through the Apify MCP server, so AI agents (Claude, ChatGPT, or anything that speaks MCP) can call it as a tool — hand it a list of entity names and read back structured facts, pageview trends and briefs — with no custom integration code.

---

Run it locally

npm install
# put your input in storage/key_value_stores/default/INPUT.json
export OPENAI_API_KEY=sk-...
apify run -p

---

Related actors

• AI Company Enrichment — Turn a list of domains into structured company profiles: firmographics, contacts, socials and ICP fit.
• Google News Scraper — Pull fresh news headlines and articles for any query, company or topic.
• B2B Account Brief — Build a ready-to-read account brief for any company for sales and research.
• Research Paper Intelligence — Search and summarize academic papers with structured metadata and AI briefs.

---

Part of the No Niner toolkit — no-code web scraping, enrichment & AI data tools, all pay-per-use. Browse the full set at noniner.io.