🌐 Wikidata Knowledge Graph & Entity Scraper

Extract structured entities, unique QIDs/PIDs, multilingual labels, semantic descriptions, aliases, and direct knowledge graph URIs from the world's largest open knowledge base: Wikidata.

Engineered for entity linking, knowledge graph construction, vector DB enrichment, and AI / LLM grounding.

---

🥊 Why This Actor Beats the Competition

Feature	🌐 This Actor (litezero)	🐢 Heavy SPARQL Queries	🔒 Proprietary Knowledge APIs
Speed	⚡ < 1.0 second / request	🐌 30–60s timeouts on complex SPARQL	⏳ 2–5 seconds with rate limits
Pricing	💰 $1.00 / 1,000 entities (Free Platform Usage)	💸 Heavy compute bill ($5–$10/run)	💳 $0.05–$0.20 per query
Simplicity	🚀 Simple keyword search with auto-linking	🧩 Requires expert SPARQL syntax	📝 Complex schema transformations
Reliability	🛡️ Official Wikimedia API with smart retries	🚫 Frequent 504 Gateway Timeouts	🔑 Strict auth tokens & quotas
Output Format	📦 Clean, standardized JSON ready for RAG	⚠️ Verbose RDF XML or raw turtle	🔒 Vendor lock-in

---

🛠️ Step-by-Step: How to Use

Method 1: Apify Web Console (No-Code)

1. Enter your search term (e.g. artificial intelligence, Quantum computing, Elon Musk, or Python).
2. Select your language (e.g. en, es, de, fr, ja).
3. Choose the entity type (Item for standard concepts or Property for relations).
4. Click Start and export results to JSON, CSV, or Excel.

Method 2: Python SDK

from apify_client import ApifyClient

client = ApifyClient("YOUR_APIFY_TOKEN")
run_input = {
    "search": "quantum computing",
    "language": "en",
    "maxResults": 15
}

run = client.actor("litezero/wikidata-entity-knowledge-scraper").call(run_input=run_input)
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(f"[{item['id']}] {item['label']}: {item['description']}")

Method 3: Node.js SDK

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_APIFY_TOKEN' });
const run = await client.actor('litezero/wikidata-entity-knowledge-scraper').call({
    search: 'machine learning',
    maxResults: 10
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(`Retrieved ${items.length} entities from Wikidata.`);

Method 4: REST API / cURL

curl "https://api.apify.com/v2/acts/litezero~wikidata-entity-knowledge-scraper/runs?token=YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"search": "reinforcement learning", "maxResults": 10}'

---

📥 Input Parameters

Parameter	Type	Required	Default	Description
search	string	Yes	"artificial intelligence"	Entity name, person, place, or concept to search.
language	string	No	"en"	Language code for labels and descriptions (e.g. en, es, fr).
type	string	No	"item"	Entity type: item (QID) or property (PID).
maxResults	integer	No	20	Max entities to return (1–50).

---

📤 Output Data Dictionary & Sample JSON

{
  "id": "Q11660",
  "title": "Q11660",
  "label": "artificial intelligence",
  "description": "intelligence of machines and software",
  "concepturi": "http://www.wikidata.org/entity/Q11660",
  "url": "https://www.wikidata.org/wiki/Q11660",
  "aliases": [
    "AI",
    "machine intelligence",
    "synthetic intelligence"
  ],
  "match_type": "label",
  "match_text": "artificial intelligence",
  "language": "en",
  "scraped_at": "2026-10-04T06:00:00.000Z"
}

---

💡 Top Commercial Use Cases

1. Entity Disambiguation & Linking: Automatically map ambiguous text mentions in unstructured documents to canonical Wikidata QIDs.
2. Knowledge Graph Generation: Extract core nodes, properties, and synonym aliases to bootstrap enterprise knowledge graphs.
3. Semantic RAG Context Injection: Ground LLM prompts with verified factual summaries and encyclopedic identifiers directly from Wikimedia.