Overview
Ento-Linguistics v1.3.1 is a corpus-custody patch to the v1.3.0 complexity edition. A second exact-text PubMed reconciliation links 59 more archived abstracts to their source records (7,599 of 7,609 now identified; the 10 unreconciled strings stay excluded from headline results), and PMC records titled as corrections, errata, retraction notices or retracted items are excluded from full-text analysis (6,997 of 7,073 stored records analyzed). All analysis layers, both fixed-margin protocols and the vocabulary/threshold diagnostic were regenerated; the paper reports every value from those exports. Six overlapping terminology domains and four source layers retain their descriptive scope; fixed-margin comparisons and heuristic entropy/CACE measures remain distinct from biological interactions, validated meaning or causal reader effects. The 20-minute lecture remains the v1.3.0 GitHub release asset matched to the v1.3.0 paper. Paper figures are CC BY 4.0; software is MIT; source corpus materials retain their providers' terms. Previous version DOIs preserve earlier artifact identities.
Overview source: Curated paper metadata.
Methods and contributions
Read the source for the full argument, qualifications, and evidence.
Findings and contributions
- The corpus (48787 tokens) yields 888 candidate terms, 261 of them assigned to domains, across 6 conceptual clusters linked by 9 weighted relationships.
- Terminology networks are strongly modular with cross-domain bridging, most prominently in Power and Labor, which has 43 bridging terms.
- 16.9% of the 261 domain-assigned terms exhibit context-dependent semantic drift.
- Economics terms have the highest mean semantic entropy (1.21 bits) of all domains despite having zero bridging terms.
- CACE scoring of the "slave" to "host worker" reform shows aggregate scores rising from 0.38 to 0.81.
Methods
- Corpus of 369 entomology publications mined from PubMed and arXiv — Literature-mining classes queried PubMed (ants/eusocial/colony terms) and arXiv q-bio, then deduplicated and quality-filtered the records.
- Seed-expansion term extraction into six Ento-Linguistic domains — Tokens matched to domain seed lexicons (e.g. Power & Labor, Kin & Relatedness, Economics) are extended to co-occurring tokens in a 3-token window.
- Semantic entropy of terms via k-means over TF-IDF usage contexts — Ambiguity of each sufficiently attested term is scored as Shannon entropy (bits) over clustered usage contexts.
- Six-layer deterministic analysis pipeline including networks and CACE scoring — Layers cover extraction, entropy, domain statistics, conceptual networks/centrality, rhetorical scoring, and CACE meta-standard evaluation.
Summary sources: Paper metadata and evidence · Extracted source text.
Document downloads
PDFs are hosted here; supporting sources may use pinned archives.
- Ento_Linguistics_DAF_TCC_v1_04-15-2026.pdf PDF · 5.62 MiB · extracted-text source
- Ento_Linguistics_manuscript.pdf PDF · 11.02 MiB
These files may be different versions or companion documents. The archive does not identify a latest edition.
Citation
Citation metadata follows the unified bibliography.
Catalog details and resources
Related works
Ranked by recorded methods, topics, contributors and associated software, with contributor roles and domain-only matches labelled. Metadata overlap is not a citation or a scientific equivalence.