Percentage of conference presenters found in the Japanese Literary Studies dissertation database. Higher = more disciplinary overlap.
Lower values = more similar topic distributions between venues.
This dashboard provides a comparative view of Japanese Literary Studies conference presence across five major academic venues:
Data spans from 1968 (earliest MLA convention program data) through 2026, with varying coverage windows per venue.
Data Sources: Conference programs harvested via Confex API, legacy HTML scraping (Wayback Machine), PDF text extraction, and PMLA/PAJLS proceedings. Japan-relevance determined by keyword filtering with human review.
Person Matching: Presenters matched across venues using normalized
name comparison with the central persons database (727K records).
Match rates reflect disciplinary proximity: AJLS (78.7%) → SCMS (21.3%).
Topic Classification: Titles classified into 15 thematic categories via keyword pattern matching against curated dictionaries. Anthropic Claude API used for NLP-enhanced classification where available.
Jensen-Shannon Divergence: Measures topic distribution similarity between venue pairs. Values range from 0 (identical) to ln(2) ≈ 0.693 (maximally different). Computed from normalized topic proportions.
HHI (Herfindahl-Hirschman Index): Measures institutional concentration across venues. Values range from 0.2 (perfectly distributed across 5 venues) to 1.0 (present at only 1 venue).