HOW TF-IDF WORKS
Term frequency and rarity are combined into one score
TF-IDF combines within-document term frequency with inverse-document frequency across the document pair. Words that appear in both texts with similar intensity keep a low score; words that are common in only one text rise in ranking.
TF
term frequency ÷ analyzed wordsIDF
log((N + 1) ÷ (df + 1)) + 1COMPARATIVE VIEW
Inspect both documents with one shared weighting model
Two-source setup
Use text or a public URL for each source and keep settings synchronized for fair comparison.
Top terms
Limit rows to a deterministic top slice to focus on the strongest terms.
Global IDF table
Review which terms carry more discriminative power across the pair.
Export
Download both document vectors and metadata for reproducible modeling.