CVV Example

1
CVV Example

2
DFi,j Assumption

• If we assume that DFi,j the number of documents
  in collection ci containing term tj
• A DFi,j / Ni
• proportion of docs in ci containing term tj
• B Sumk!iC(DFk,j) / Sumk!iC(Nk)
• proportion of docs not in ci containing term tj
• not the same as Sumk!iC (DFk,j/Nk)
• A B ! proportion of all docs containing tj
• see example on next page

3
CVV Example (for one term tj)

• Given C 3, DF1..3,j1,2,0, N1..32,4,4
• c1 A1/2, B (20) / (44) 2/8 1/4
  AB 1/2 1/4 3/4, CV1,j (1/2)/(3/4) 2/3
• prop of all docs containing tj (12) / (244)
  3/10
• c2 A 2/41/2, B (10) / (24) 1/6
  CV2,j (1/2) / (1/2 1/6) (1/2) / (4/6) 3/4
• c3 A 0/4 0, B (12) / (24) 3/6 1/2
  CV3,j 0 / (0 1/2) 0
• So, CV1..3,j 2/3, 3/4, 0

4
CVV Example (cont)

• CV1..3,j 2/3, 3/4, 0 from previous page
• avgCVj Sumi1C(CVi,j) / C
  (2/3 3/4 0) / 3 .472
• CVVj Sumi1C(CVi,j - avgCVj)2 / C
  ((.667-.472)2 (.75-.472)2 (0-.472)2) /3
  (.0378 .0773 .2228) / 3
  .113

5
CVV Example (cont)

• CVVj.113, DFi,j1,2,0 from previous pages
• Given query q has only one term in query tj
  (M1)
• Gi,q Sumk1M(CVVk DFi,k) CVVj
  DFi,j for our example
• G1..3,q .113, .226, 0
• So, collection c2 is gooder than the others...
• Goodness is only an indicator as to where, among
  the C collections, the query terms are
  concentrated at. lt-- bad grammar!