Skip to content

Interface Overview

This guide covers the main interface elements and navigation patterns in ProEthica.

Access Levels

ProEthica operates in demo mode by default, allowing exploration of cases and extracted entities without authentication. The interface adapts based on access level:

Level Access Available Features
Anonymous (Demo) No login required Browse cases, view extractions, explore precedent network, read guidelines
Authenticated Login required Run extractions, edit entities, manage pipeline queue
Admin Admin account User management, domain configuration, system settings

Users exploring the demo will see view-only interfaces. Buttons for extraction, editing, and pipeline operations appear only for authenticated users. This documentation describes the full interface as seen by authenticated users.

The top navigation provides access to main features:

Menu Item Type Contents
Home Link Return to home page
[Domain] Dropdown Current domain (e.g., Engineering)
Cases Link Case repository
Case Relations Dropdown Similar Cases, Similarity Network, Lineage Graph
Guidelines Link Browse ethical guidelines and codes of ethics
Docs Link This documentation
Tools Dropdown Academic References; Linked Data (OntServe Web, Browse ProEthica Ontologies); Admin Tools (Batch Processing, Batch Queue, Validation Studies, Prompt Editor)
[User] Dropdown User menu with Logout

Case Relations Dropdown

Tools Dropdown

The domain dropdown (e.g., "Engineering") reflects the currently active professional domain. A "Development" badge appears in development mode with full access enabled without authentication.

Cases Section

Case Repository

The Cases page (/cases/) displays all uploaded cases with:

  • Year grouping for chronological organization
  • Subject tags for filtering by topic
  • Question and conclusion previews (expandable/collapsible via view toggle)
  • Compact and expanded view modes (persisted in browser storage)

Case Repository

Case Detail

Each case has a detail page showing:

  • Case title, reference number, and metadata
  • Context bar with OntServe entity link, transformation badge, and utility buttons (Structure, Provenance)
  • Full narrative with Facts and Discussion sections
  • Interactive RDF triple labels linking to related cases
  • Pipeline status bar and step buttons (visible to authenticated users only in production)

Case Detail

Document Structure

The Structure view (/cases/<id>/structure) provides:

  • Section breakdown with character counts
  • Embedding status and regeneration controls
  • Similar cases based on semantic matching

Case Structure

Extraction Pipeline

The pipeline provides step-by-step case analysis (requires authentication). Access it from the case detail page via the Pipeline button or the per-case pipeline dashboard at /cases/<id>/pipeline.

Step Name Description Entity Types
1 Contextual Framework Extract from Facts and Discussion sections Roles (R), States (S), Resources (Rs)
2 Normative Requirements Extract from Facts and Discussion sections Principles (P), Obligations (O), Constraints (Cs), Capabilities (Ca)
3 Temporal Dynamics Unified extraction via LangGraph Actions (A), Events (E), causal chains, Allen relations
Reconcile Entity Deduplication Merge overlapping entities across passes --
4 Whole-Case Synthesis Multi-phase analysis (2A-2E, Phase 3, Phase 4) Provisions, Precedents, Questions, Conclusions, Decision Points, Resolution Patterns, Causal-Normative Links, Question Emergence

Pipeline Sidebar

The sidebar shows all pipeline steps with checkmarks for completed steps. Step 4 has sub-views: Extraction, Review, and Full View.

Pipeline Overview

Step Views

Each step displays extracted entities organized by concept type with:

  • Facts/Discussion toggle (Steps 1-2) for section-specific extraction
  • Entity counts showing classes and individuals per concept
  • Re-run Extraction button to re-extract
  • Provenance link to view extraction history
  • Draft/Committed status badge

Step 1 Extraction

Step 4 Full View

The Full View provides a tabbed interface with 8 tabs: Entities (interactive graph), Flow, Provisions, Precedents, Q&C, Analysis, Decisions, and Narrative.

Step 4 Full View

Precedent Discovery

The Precedents page (/cases/precedents/) provides:

  • Similarity scores based on document embeddings
  • Breakdown by section (Facts, Discussion)
  • Filter by specific case or view all pairs

Precedent Discovery

Similarity Network

The network view (/cases/precedents/network) visualizes case relationships:

  • Node color indicates case outcome (green: ethical, red: unethical, orange: mixed)
  • Edge color indicates similarity strength
  • Click nodes for case details
  • Click edges for similarity breakdown

Similarity Network

Pipeline Automation

For batch processing, Tools > Batch Processing (/pipeline/dashboard) shows:

  • Service status (Redis, Celery, queue depth)
  • Active pipeline runs with progress bars
  • Queue management controls

Pipeline Dashboard

Status Indicators

Service Status

The header shows service connectivity:

  • Green - All services operational
  • Yellow - Some services unavailable
  • Red - Critical services offline

Entity Status

Entities in review display their state:

  • New - Extracted this session
  • Existing - Matches ontology class
  • Modified - User-edited entity

Keyboard Shortcuts

Shortcut Action
Ctrl+Enter Submit current form
Esc Close modal dialogs