CRANV2 Aslib Cranfield Research Project: Factors Determining the Performance of Indexing Systems: Volume 2 Simulated ranking and document output cut-off 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. - 204 - SMART CRANFIELD NORMALISED SMART NORMALISED RECALL AND LANGUAGE RECALL PRECISION SI 58.64 1.492 $2 61.06 1.546 $3 62.70 1.573 $4 58.58 1.495 $5 62,41 1.573 $6 64.88 1.609 $7 61.82 1.548 $8 63.64 1.594 $9 65.13 1.618 S10 62.94 1.579 Sll 64.94 1.617 S12 63.64 1.593 S13 65.23 1.624 S14 64.82 1.612 FIGURE 5.9T PERFORMANCE FIGURES FOR SMART LANGUAGES ORDER CRANFIELD SMART 1 S13 S13 2 $9 $9 3 SII $11 4 $6 S14 5 S14 $6 6 $8 $8 7 S12 S12 8 SI0 SI0 9 $3 $3 10 $5 $5 11 $7 $7 12 $2 $2 13 S1 $4 14 $4 SI FIGURE. 5.10T COMPARISON OF RANKING OF SMART LANGUAGES BY CRANFIELD AND SMART NORMALISED MEASURES