taxonomy-build

You build a taxonomy by organizing paper analysis results into a hierarchy of method category, submethod, and specific technique. This generates a machine-readable taxonomy.md with evidence-grounded nodes and cross-references for survey construction.

A machine-readable taxonomy for survey construction is a hierarchical classification of papers by training paradigm and evaluation focus. It structures paper analysis results into export-ready formatting with cross-references and coverage reporting in taxonomy.md.

You classify research papers by processing their analysis results into a Level 1 method category, Level 2 submethod, and Level 3 specific technique. This generates evidence-grounded taxonomy nodes with cross-references.

Yes, generating a hierarchical taxonomy requires prior paper analysis results. You must place existing analysis outputs into the paper_analysis_results/ directory to process and generate the final taxonomy.md file.

Yes, you can use the generated taxonomy for downstream gap analysis. The taxonomy.md provides lossless integration with gap-identify workflows, supporting automated synthesis and coverage reporting for your research survey.

The best way to ensure evidence-grounded taxonomy nodes is to generate them from structured paper analysis results. This process binds evidence to hierarchical classifications and produces cross-references with coverage reporting in taxonomy.md.