Skip to content

First Login

This guide covers your first login and navigating the OntExtract interface.

Logging In

Docker Installation

If you used Docker, a default admin account is created automatically:

  • Username: admin
  • Password: admin123

Navigate to http://localhost:8765 and click Login in the top navigation.

Login Page

Creating a New Account

If you need to create your own account:

  1. Click Register in the top navigation
  2. Enter your username, email, and password
  3. Click Create Account

Interface Overview

After logging in, you'll see the home page with the research workflow.

Home Page

The top navigation bar provides access to:

Menu Item Description
Library Sources (documents and references) and Anchor Terms
Experiments Your analysis experiments
Provenance Timeline and Graph views of processing history (dropdown menu)
Linked Data Ontology and semantic resources
Docs This documentation

Quick Action Buttons

The home page displays quick action buttons:

  • Add Term - Create a new anchor term
  • Add Source - Upload a document
  • Experiments - View or create experiments
  • Results - Access analysis results

Research Workflow

The 6-step workflow cards guide you through the analysis process:

  1. Define Terms - Create anchor terms to track
  2. Upload Sources - Add documents from different periods
  3. Create Experiment - Link terms to documents
  4. LLM Orchestration - AI suggests processing strategies
  5. Execute Pipeline - Process documents with selected tools
  6. View Results - Explore findings and provenance

Operational Modes

OntExtract operates in two modes depending on your configuration:

Standalone Mode

Available without any API keys. Features include:

  • Manual tool selection
  • Entity extraction (spaCy)
  • Temporal analysis (NLTK)
  • Text segmentation
  • Embedding generation (sentence-transformers)
  • PROV-O provenance tracking

API-Enhanced Mode

Enabled when an Anthropic API key is configured. Additional features:

  • LLM-powered tool recommendations
  • Automated document analysis
  • Cross-document synthesis
  • LLM text cleanup for OCR errors

To enable API-enhanced mode, add your API key in Settings or via environment variable.

1. Explore the Sample Experiment

If available, browse the pre-loaded "Agent Temporal Evolution" experiment to see:

  • How documents are organized by period
  • Processing artifacts generated
  • Provenance tracking in action

2. Upload a Test Document

  1. Click Add Source or navigate to Library > Sources
  2. Upload a PDF or text file
  3. Review the automatic metadata extraction

3. Create an Anchor Term

  1. Navigate to Library > Terms
  2. Click Add New Anchor Term
  3. Enter a term like "agent" or "intelligence"
  4. Assign a domain (e.g., "Artificial Intelligence")

4. Create Your First Experiment

  1. Go to Experiments > New Experiment
  2. Enter a name and description
  3. Select documents to include
  4. Choose your anchor terms
  5. Save and start processing

Getting Help

  • Documentation: Click Docs in the navigation
  • FAQ: Check the FAQ for common questions
  • Issues: Report problems on GitHub

Next Steps