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. - 188 - FIGURE 4.902T Index Language IV. 3. a (Single terms. Natural language. Exhaustivity of Indexing Title and Abstract Search Rule A Document Relevance I - 4 Number of Documents in Collection 200 (Subset 1) Number of Questions 42 (Subset 2} Number of Relevant Documents 198 GeneraLity Number 23.6 Coordination) Coord- Documents Recall Precision FaUout ination Retrieved Ratio Ratio Ratio x y z Level Rel Non-rel. a/a+c a]a+b b]b+d 1 185 6,261 93.4% 2.9% 76.353% 42 0 42 2 lfil 2,158 61.3% 6.9% 26.311% 42 42 42 3 120 854 60.6% 12.3% 10.412% 40 42 42 4 75 288 37.9% 20.7% 3.511% 32 41 41 5 44 97 22.2% 31.2% 1.183% 21 39 39 6 27 25 13.6% 51.9% 0.305% 13 33 33 7 12 2 6.1% 85.7% 0. 024% 6 27 27 8 3 0 1.5% IOO. 0% o. 000% 2 16 16 9 2 0 1.0% 100.0% O. 000% 1 11 11 1O 0 0 0 7 7 11 0 0 0 3 3 12 0 0 0 1 1 FIGURE 4.903T Index Language IV.4.a (Single .terms. Word forms. Coordination) Exhaustivity of Indexing Title and Abstract Search Rule A Document Relevance 1 - 4 Number of Documents in CoLlection 200 (Subset 1) Number of Questions 42 (Subset 2) Number of Relevant Documents 198 Generality Number 23.6 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 1 187 6,770 94.4% 2.7% 82.540% 42 0 42 2 166 2,654 83.8% 5.9% 32.358% 42 42 42 3 130 1,150 65.7% 10.2% 14.021% 41 42 42 4 92 432 46.5% 17.6% 5.267% 35 41 41 5 51 158 25.8% 24.4% 1.926% 28 39 39 6 32 53 16.2% 37.6% 0.646% 19 33 33 7 20 13 10.1% 60.0% 0.158% 10 27 27 8 9 4 4.5% 69.2% o.o49% 7 18 18 9 3 1 1.5% 75.0% 0. 012% 4 11 11 10 I 0 lOO.O% o.ooo[OCRerr] 1 7 7 11 0 0 0.5% 0 3 3 12 0 0 0 1 1