Datasets:

Corp-o-Rate-Community
/

entity-references

license: mit
task_categories:
  - text-classification
  - feature-extraction
language:
  - en
tags:
  - entity-resolution
  - named-entity-recognition
  - company-data
  - embeddings
  - sqlite
  - vector-search
size_categories:
  - 1M<n<10M

Entity References Database

A pre-built SQLite database with vector embeddings for organization/entity resolution. Used by the corp-extractor library for fast entity qualification via embedding similarity search.

Overview

This database contains organization records from multiple authoritative sources, each with:

Organization name (canonical)
Source identifier (LEI, CIK, UK Company Number, Wikidata QID)
Entity type classification (business, nonprofit, government, etc.)
Vector embeddings for semantic search (768-dim, using embeddinggemma-300m)

Data Sources

Source	Records	Identifier	Description
GLEIF	~3.2M	LEI (Legal Entity Identifier)	Global legal entities from the LEI system
SEC Edgar	~100K+	CIK (Central Index Key)	US SEC-registered filers
Companies House	~5M	UK Company Number	UK registered companies
Wikidata	Variable	Wikidata QID	Notable organizations from Wikidata

Entity Types

Organizations are classified into the following types:

Category	Types
Business	`business`, `fund`, `branch`
Non-profit	`nonprofit`, `ngo`, `foundation`, `trade_union`
Government	`government`, `international_org`, `political_party`
Other	`educational`, `research`, `healthcare`, `media`, `sports`, `religious`, `unknown`

Database Variants

File	Description	Use Case
`entities.db`	Full database with complete source record metadata	When you need full record details
`entities-lite.db`	Lite version without record data	Default - faster download, smaller size
`entities.db.gz`	Compressed full database	When bandwidth is limited
`entities-lite.db.gz`	Compressed lite database	Smallest download size

Schema

organizations table

CREATE TABLE organizations (
    id INTEGER PRIMARY KEY AUTOINCREMENT,
    name TEXT NOT NULL,
    name_normalized TEXT NOT NULL,
    source TEXT NOT NULL,  -- 'gleif', 'sec_edgar', 'companies_house', 'wikipedia'
    source_id TEXT NOT NULL,
    region TEXT NOT NULL DEFAULT '',
    entity_type TEXT NOT NULL DEFAULT 'unknown',
    record TEXT NOT NULL,  -- JSON with full source record (empty in lite version)
    UNIQUE(source, source_id)
);

organization_embeddings table (sqlite-vec)

CREATE VIRTUAL TABLE organization_embeddings USING vec0(
    org_id INTEGER PRIMARY KEY,
    embedding float[768]
);

Usage with corp-extractor

# Install
pip install corp-extractor

# Download the database (lite version by default)
corp-extractor db download

# Download full version
corp-extractor db download --full

# Search for an organization
corp-extractor db search "Microsoft"

# Check database status
corp-extractor db status

Python API

from statement_extractor.database import OrganizationDatabase, CompanyEmbedder

# Load database
database = OrganizationDatabase()
embedder = CompanyEmbedder()

# Search by embedding similarity
query_embedding = embedder.embed("Microsoft Corporation")
results = database.search(query_embedding, top_k=5)

for record, similarity in results:
    print(f"{record.name} ({record.source}:{record.source_id}) - {similarity:.3f}")

Building Your Own Database

# Import from authoritative sources
corp-extractor db import-gleif --download
corp-extractor db import-sec --download
corp-extractor db import-companies-house --download
corp-extractor db import-wikidata --limit 50000

# Upload to HuggingFace
export HF_TOKEN="hf_..."
corp-extractor db upload

License

MIT License - the database structure and embedding generation code are MIT licensed.

Individual data sources have their own licenses:

GLEIF: Open license for LEI data
SEC Edgar: Public domain (US government)
Companies House: Open Government Licence
Wikidata: CC0 (public domain)