Comp4321 The Index May 14 2026

The anatomy of a top result.

query · postgradu admiss program comput applic · doc № 006
body match

Forward index · top stems on this page

program 16 comput 13 student 7 univers 6 scienc 6 engin 6 studi 6 depart 5 hkust 5 train 5 graduat 5 year 4 inform 4 also 4

document length 288 body tokens  ·  max tf 16 (program)  ·  vocabulary 163 distinct stems  ·  title 1 stems

Score breakdown

Re-derived from the same tf·idf cosine values, boost factors, and PageRank multipliers the live ranker just used. weighted = body + title × 2.0 · final = weighted × phrase × pagerank.

  1. cosine(query, body) tf·idf body · 5 resolved stems0.3943
  2. cosine(query, title) × 2.0 no title overlap (×2.0 unused)0.0000
  3. weighted subtotal cosBody + 2.0 × cosTitle0.3943
  4. × phrase multiplier stems not within 5-token window · ×1.0×1.0
  5. × pagerank multiplier 1 + 0.5 × normalized PR (0.000)×1.0002
  6. final score weighted × phrase × pagerank0.3943

Per-query-stem · how each term contributes

  1. postgradu wid 0676 · df 2 body tf 1 title tf 0 idf 7.229
  2. admiss wid 0004 · df 3 body tf 0 title tf 0 idf 6.644
  3. program wid 0674 · df 5 body tf 16 title tf 0 idf 5.907
  4. comput wid 0683 · df 7 body tf 13 title tf 0 idf 5.421
  5. applic wid 0675 · df 3 body tf 1 title tf 0 idf 6.644

Outlinks (1)