CRANV2 Aslib Cranfield Research Project: Factors Determining the Performance of Indexing Systems: Volume 2 Main test results chapter Cyril Cleverdon Michael Keen Cranfield An investigation supported by a grant to Aslib by the National Science Foundation. Use, reproduction, or publication, in whole or in part, is permitted for any purpose of the United States Government. - 88 - FIGURE 4. 102T Index LanguaEe I. 3. a (S. T. Word forms, Exhaustivity of Indexing 3 Search Rule A Document Relevance I - 4 Number of Documents in Collection i. 400 Number of Questions 221 (Subset 3} Number of Relevant Documents 1. 590 Generality Number 5.1 Coordination) Coord- Documents Recall Precision Fallout ination Retrieved Ratio Ratio Ratio x y z Level Rel. Non-rel. a/a+c a]a+b b[b+d 1 1,533 (-) 96.4% (-) (-) 221 O 221 2 1,338 62,765* 84.2% 2. i%* 20. 3[OCRerr] 8[OCRerr]/˘* 221 44* 221 3 1,017 24,726* 64.0% 3.9%* 8. 028%* 217 109" 220 4 677 9, 565* 42.6% 6.6%* 2. 530%* 192 142" 212 5 374 3, 084. 23.5% 10. 8%* 1. 001%* 139 177" 197 6 192 1,112. 12.1% 14. 8%* 0. 361%* 99 164" 161 7 96 333 6.0% 22.4% 0.108% 64 140 140 8 34 87 2.1% 28.1% 0.028% 28 105 105 9 13 15 0.8% 46.4% 0.005% 17 78 78 I0 2 0 o,1% lOO. O% 0.000% 2 52 52 II 0 O 0 32 32 12 0 0 0 15 15 13 0 0 0 8 8 14 0 0 0 4 4 15 0 0 0 3 3 FIGURE 4. 103T Index Language L 5. a (S. T. Synonyms, Quasi-synonyms. Coordination) Exhaustivity of Indexing 3 Search Rule A Document Relevance 1 - 4 Number of Documents in Collection 1,400 Number of Questions 221 (Subset 3) Number of Relevant Documents i, 590 Generality Number 5. i Coord- Documents Reca[OCRerr] Precision Fallout Łaation Retrieved Ratio Ratio Ratio x y z Level Rel. Non-tel. a/a+c a/a+b b]b+d 1 1,548 (-) 97.4% (-) 1.. [OCRerr]%* (-) 221 0 221 2 1,406 114,265" 88.4% 37. 099%* 221 44[OCRerr] 221 3 1,121 42,364* 70.5% 2.6%* 13.755%* 218 109" 220 4 802 16,191" 50.4% 4.77[OCRerr] 5, 257%* 204 142[OCRerr] 212 5 475 8,164" 29.9% 5. 5%* 8. 1%* 2.651%* 164 177" 197 6 265 3,013. 16.7% 0. 278%* 114 161" 164 7 131 910 8.2% 12.6% 0.296% 79 140 140 8 50 266 3.1% 15.8% 0.088% 44 105 105 9 19 56 1.2% 25.3% o. o18% 20 78 78 i0 2 12 0.1% 14.3% 0.004% 6 52 52 11 0 O 0 32 32 12 0 0 0 15 15 13 O O 0 8 8 14 0 0 0 4 4 15 0 0 0 3 3