This folder contains the current SURA data, embeddings, analysis outputs, university-ranking crosswalk, final case files, final analysis report, and the scripts used for the final processing steps.
data/— current source and prepared datasets.embeddings/— current text embeddings, metadata, and unit vectors.exports/— flat embedding exports.results/— final similarity matrices, D1–D9 outputs, audits, statistics, plots, and the final university-ranking workbook.deliverables/0707/— final panel, program, outcome, and case outputs.deliverables/0716/— final inverse-case outputs and ranking crosswalk tables.reports/— final exploratory analysis report.scripts/— final processing and ranking scripts only.
Start with OFFBOARDING.md for the final output map and execution pipeline. The complete file inventory is in offboarding_upload_manifest.csv.
results/University Rankings - Master Database.xlsxresults/D1_density_overlay_pb_vs_cb.pngresults/D2_bizidea_top3_pitchbook.csvresults/D3_bizidea_top3_crunchbase.csvresults/D4_insider_outsider_table.csvresults/D5_vigintile_cutoffs_per_industry.csvresults/D6_bizidea_evolution_density.pngresults/D7_team_bizidea_evolution.csvresults/D8_bizidea_all_nolabels_vs_industry.csvresults/D9_weekly_density_labeled_vs_unlabeled.pngreports/exploratory_analysis_report.md
The package excludes superseded duplicate evolution files, earlier top-three files, dated administrative materials, and previous pipeline versions.