Data & Replication

All data, scripts, and outputs are available in the project repository for full replication.

Repository

Key Data Files

File	Description
output/data/channel_rankings.json	Composite scores, sub-scores, and ranks for all 14 candidate channels (unscreened corpus)
output/data/channel_rankings_filtered.json	Rankings recomputed after removing off-topic papers identified by relevance screening
output/data/sensitivity_analysis.json	Four alternative weight schemes with per-channel composite scores and ranks
output/data/filtered_vs_unfiltered_comparison.json	Side-by-side comparison of filtered and unfiltered rankings with rank changes
output/data/corpus_contamination_audit.json	Audit of off-topic papers per channel: counts removed, reasons, and papers remaining
output/data/openalex_merged.json	Merged OpenAlex corpus with all 2,433 papers, metadata, and channel assignments
output/data/per_paper_scoring_detail.json	Per-paper scoring breakdown: relevance flags, citation counts, channel membership
output/data/key_papers.json	Curated list of high-impact papers per channel used for narrative synthesis

Key Scripts

Script	Purpose
scripts/channel_mapper.py	Computes composite scores and rankings for all 14 channels from the unscreened corpus
scripts/channel_mapper_filtered.py	Recomputes rankings after removing off-topic papers; produces filtered comparison
scripts/openalex_search.py	Queries OpenAlex API with Boolean keyword expressions; retrieves and deduplicates papers
scripts/sensitivity_analysis.py	Tests four weight schemes and reports rank stability across configurations
scripts/verify_all_papers.py	Validates every paper in the corpus against OpenAlex metadata (DOI, title, year)
scripts/hostile_review.py	Automated adversarial review: checks for citation errors, data inconsistencies, and methodology gaps
scripts/reference_audit.py	Cross-references every in-text citation against the bibliography; flags orphans and phantoms

Paper Downloads

Citation