56% of relevant papers only mention search terms in full-text, not abstracts
Traditional Search
Title + Abstract Only
2,821
44%
Papers Found
OpenAlex Full-Text
Title + Abstract + Full-Text
6,329
100%
Papers Found
Hidden Literature
3,508
56% of Relevant Papers
Only mentioned in methodology, case studies,
comparative analyses, or body text
Why Are They Hidden?
📊 Comparative Studies
Papers comparing multiple countries mention Russian policy in analysis sections
🔬 Methodology Papers
Research methods papers using Russian policy as example cases
📚 Historical Analyses
Broader historical works with sections on Russian policy
🌍 Regional Studies
Area studies mentioning Russian policy in regional context
📖 Book Chapters
Edited volumes with Russian policy discussed in specific chapters
The Zotero Challenge
⚠️ Traditional Zotero Import Limitations
🔍
Browser Extension
Only sees visible results
📑
Metadata Matching
Relies on titles/abstracts
❌
Result
Miss 56% of papers
✅ OpenAlex → Zotero Solution
🌐
API Export
Get ALL results
📄
RIS/BibTeX Format
Convert to Zotero format
✨
Result
Capture 100% of papers
Why We Need Everything: Dual Purpose
📚
Reference Management
✓ Citations & footnotes
✓ Bibliographies
✓ Literature reviews
✓ Academic writing
🤖
LLM Corpus Analysis
✓ Full-text processing
✓ Pattern discovery
✓ Taxonomic annotation
✓ Knowledge graphs
Precision × Recall = Comprehensive Corpus
Maximum relevant papers × Minimum noise = Optimal LLM training data
Next Workshop Sessions: LLM-Powered Analysis
1
TODAY: OpenAlex & Zotero Integration
• Complete corpus collection (6,196 papers)
• Dual-purpose workflow for reference management
• Export to RIS/BibTeX for LLM analysis
2
Building Rich Taxonomies with LLMs
• Multi-level classification systems
• Multiple perspectives (theoretical, methodological, thematic)
• Emergent categories from 6,196 papers
• LLM-suggested hierarchies
3
Intelligent Corpus Annotation
• Chunk documents into segments
• Apply taxonomic labels via LLMs
• Create knowledge graphs
• Discover hidden connections
From 6,196 papers →
Structured knowledge
Including the 3,508 "hidden" papers only findable via full-text search
Workshop Resources & Materials
Everything you need in one place
📁 Dropbox Folder: /2603 - Boston/
📂 Main Workshop Materials
├── 📊 260311-13 Part 1 - Setting the Scene.pptx (117MB)
├── 📊 260311-13 Part 2 - Building Corpora.pptx (74MB)
├── 📊 260311-13 Part 3 - Building Taxonomies.pptx (22MB)
└── 📊 260311-13 Part 4 - Using LLMs.pptx (33MB)
