1. Phonetic Support Architecture for WHG

• 1.1. Overview
  • 1.1.1. Goals
  • 1.1.2. Why Not Elasticsearch’s Built-in Phonetic Analysis?
  • 1.1.3. Limitations
  • 1.1.4. Architecture Summary
  • 1.1.5. Data Sources
• 1.2. Elastic Management Guide
  • 1.2.1. Table of Contents
  • 1.2.2. Architecture Overview
  • 1.2.3. Installation
  • 1.2.4. Configuration
  • 1.2.5. Storage Architecture
  • 1.2.6. Production Instance (VM)
  • 1.2.7. Staging Instance (Slurm)
  • 1.2.8. Authority Data Ingestion
  • 1.2.9. Index Management
  • 1.2.10. Snapshot Management
  • 1.2.11. Production Deployment
  • 1.2.12. Health Monitoring
  • 1.2.13. Troubleshooting
  • 1.2.14. Quick Reference
• 1.3. Infrastructure Summary
  • 1.3.1. Overview
  • 1.3.2. Architecture
  • 1.3.3. Authority Data Sources
  • 1.3.4. WHG-Contributed Datasets
  • 1.3.5. Phonetic Search
  • 1.3.6. Index Schemas
  • 1.3.7. Deployment Strategy
  • 1.3.8. Storage Requirements
  • 1.3.9. Snapshot Strategy
  • 1.3.10. Resource Summary
  • 1.3.11. Operational Commands
  • 1.3.12. Directory Structure
  • 1.3.13. References
• 1.4. Components
  • 1.4.1. Infrastructure
  • 1.4.2. Elasticsearch Indices
  • 1.4.3. Processing Components
• 1.5. Data Flow
  • 1.5.1. Authority File Ingestion
  • 1.5.2. WHG-Contributed Dataset Ingestion
  • 1.5.3. Embedding Generation
  • 1.5.4. Incremental Updates
• 1.6. Elasticsearch Index Design
  • 1.6.1. Index Schemas
  • 1.6.2. Ingest Pipelines
  • 1.6.3. HNSW Configuration
  • 1.6.4. Analysers
• 1.7. Training the Phonetic Similarity Model
  • 1.7.1. 1. Architecture Overview
  • 1.7.2. 2. Key Features
  • 1.7.3. 3. Installation
  • 1.7.4. 4. The Training Pipeline
  • 1.7.5. 5. Python Inference API
  • 1.7.6. 6. Configuration
• 1.8. Query Pipeline
  • 1.8.1. Search Strategy
  • 1.8.2. Elasticsearch Query Structure
  • 1.8.3. Error Handling
  • 1.8.4. Performance Optimisation
• 1.9. Advantages of This Architecture
  • 1.9.1. Two-Instance Isolation
  • 1.9.2. Unified Infrastructure
  • 1.9.3. Scalable Storage Tiers
  • 1.9.4. Zero-Downtime Deployments
  • 1.9.5. Flexible Embedding Generation
  • 1.9.6. Toponym Deduplication
  • 1.9.7. Efficient Phonetic Search
  • 1.9.8. Graceful Degradation
  • 1.9.9. Unified Search Across Sources
  • 1.9.10. Maintainability
  • 1.9.11. Research Reproducibility
• 1.10. Monitoring & Observability
  • 1.10.1. Key Metrics
  • 1.10.2. Health Check Endpoints
  • 1.10.3. Log Locations
  • 1.10.4. Dashboards
  • 1.10.5. Alerting
  • 1.10.6. Runbook: Common Issues
• 1.11. Future Extensions
  • 1.11.1. Short-Term Enhancements
  • 1.11.2. Medium-Term Development
  • 1.11.3. Long-Term Vision
• 1.12. Deployment Plan
  • 1.12.1. Phase 1: Infrastructure Setup
  • 1.12.2. Phase 2: Core Index Population
  • 1.12.3. Phase 3: Model Training
  • 1.12.4. Phase 4: Embedding Generation
  • 1.12.5. Phase 5: Query Integration
  • 1.12.6. Phase 6: Production Rollout
  • 1.12.7. Ongoing Operations
• 1.13. Risk Assessment
  • 1.13.1. Technical Risks
  • 1.13.2. Operational Risks
  • 1.13.3. Data Quality Risks
  • 1.13.4. Mitigation Strategies
• 1.14. Success Criteria
  • 1.14.1. Technical Metrics
  • 1.14.2. User Experience Metrics
  • 1.14.3. Research Impact
  • 1.14.4. Acceptance Criteria by Phase
• 1.15. Summary & References
  • 1.15.1. Key Design Decisions
  • 1.15.2. Technology Stack
  • 1.15.3. Storage Requirements
  • 1.15.4. Timeline
  • 1.15.5. References