Research infrastructure
Area 01 · Systems that organize evidence and retrieval
Chapter 01
Medical vocabularies
Controlled terms, identifiers, mappings, and hierarchy rules used across biomedical evidence.
MeSH Descriptors and Supplementary Concepts: Record Types and Retrieval Roles Formal MeSH descriptors define primary answers. Supplementary concepts provide mapped supporting detail. RI.MV.001Lexical MeSH Linking: Phrase Windows and Association Evidence Deterministic phrase windows connect query text to exact vocabulary names and observed MeSH associations. RI.MV.002MeSH Hierarchy Ranking: GBM Features and Poincaré Coordinates Explicit hierarchy features and hyperbolic coordinates provide separate structural ranking evidence. RI.MV.003Clinical Text to Exact MeSH: Five Independent Embedding Models Five complete models map identical clinical records to ranked exact MeSH identifiers. RI.MV.004MeSH Ranking Feature Signals: Fifteen Additional Separators Fifteen ranking signals separate candidates that existing lexical and semantic scores leave ambiguous. RI.MV.005MeSH Training Matrix: Building One 53-Column Candidate Row One stable query-candidate row receives 53 features from shared training and inference producers. RI.MV.006Exact MeSH Linking: Qwen Embedding and Reranking Wide embedding retrieval and deep pairwise reranking produce one exact MeSH identifier. RI.MV.007
Chapter 02
Evidence pipelines
How new sources are stored, organized, and searched.
Evidence Source Routing: Where New Data Goes Original files stay in an archive. Parsed records go to the database shape that fits them. RI.EP.001Evidence Retrieval: From Original Sources to Ranked Answers An archive holds the originals, a database holds the records, and a search index ranks matches back to them. RI.EP.002
Chapter 03
Research sessions
How a research session takes questions and produces a cited report.
Chapter 04
Knowledge graphs
Entity relationships, graph lenses, traversal rules, and source traceability.