# FileAnalysis.org

> Document search and analysis platform with AI-powered Named Entity Recognition

FileAnalysis.org provides full-text search across public legal document collections with automatic extraction of people, organizations, locations, dates, and other entities.

## Site Purpose

This is a research platform for analyzing public court documents and legal filings. The site enables:
- Full-text search across document collections
- Named Entity Recognition (NER) for people, organizations, locations, dates, money, events
- Document browsing with metadata and content viewing
- Entity exploration and co-occurrence analysis
- Timeline visualization of dated documents
- Geographic mapping of location entities

## Available Document Collections

The Epstein Files: Court filings, depositions, and documents from the Jeffrey Epstein case
- Subdomain: https://epstein.fileanalysis.org
- Contains: Flight logs, court records, legal filings, associated documents

## API Access

The site provides a REST API for programmatic access. Base URL: https://fileanalysis.org/api-proxy.php

### Endpoints

GET /api/search?q={query}&index={index}
- Full-text search across documents
- Parameters: q (search query), index (collection name or "all")
- Returns: Matching documents with highlights and metadata

GET /api/documents?index={index}&page={n}&limit={n}
- Paginated document listing
- Returns: Document metadata, titles, dates, entity counts

GET /api/document/{id}?index={index}
- Single document with full content
- Returns: Complete document text, metadata, extracted entities

GET /api/entities?type={type}&index={index}
- Aggregated named entities
- Types: person, organization, location, date, money, event
- Returns: Entity names with document mention counts

GET /api/entity/{name}?index={index}
- Documents mentioning a specific entity
- Returns: List of documents containing the entity

GET /api/co-occurrence?terms={entity1,entity2}&index={index}
- Entity co-occurrence analysis
- Returns: Documents where multiple entities appear together

GET /api/locations?index={index}
- Location entities with geocoding
- Returns: Location names, coordinates, document counts

GET /api/stats?index={index}
- Collection statistics
- Returns: Document counts, entity counts, date ranges

## Content Structure

### Main Application (SPA)
- / - Dashboard with search, stats, and navigation
- /#search?q={query} - Search results
- /#documents - Document browser
- /#document/{id} - Single document view
- /#entities/{type} - Entity explorer (people, organizations, locations)
- /#entity/{name} - Entity detail page
- /#timeline - Date-based document timeline
- /#map - Geographic visualization

### SEO Pages (Static)
- /people/{slug}.php - Person entity pages
- /locations/{slug}.php - Location entity pages
- /organizations/{slug}.php - Organization entity pages
- /documents/{slug}.php - Document pages

### Legal Pages
- /terms.html - Terms of Service
- /privacy.html - Privacy Policy
- /cookies.html - Cookie Policy
- /disclaimer.html - Legal Disclaimer
- /takedown.html - DMCA/Takedown Policy
- /contact.html - Contact Information
- /accessibility.html - Accessibility Statement

## Data Format

Documents include:
- title: Document title
- content: Full text content
- date: Document date (when available)
- source: Origin/collection name
- entities: Extracted named entities by type
- metadata: Additional document attributes

Entities include:
- name: Entity text as found in documents
- type: person, organization, location, date, money, event
- count: Number of document mentions
- documents: List of document IDs containing the entity

## Usage Guidelines

- All documents are from public sources (court filings, public records)
- Content is provided for research and educational purposes
- No authentication required for read access
- Rate limiting may apply to API endpoints
- Attribution appreciated but not required

## Contact

For questions, takedown requests, or data inquiries:
- Email: support@fileanalysis.org
- Contact page: https://fileanalysis.org/contact.html
- Takedown requests: https://fileanalysis.org/takedown.html