IRS13 An Analysis of the Documentation Requests chapter E. M. Keen Harvard University Gerard Salton Use, reproduction, or publication, in whole or in part, is permitted for any purpose of the United States Government. X-31 RELEVANT RANl[OCRerr] OF RELEVANT AND RE[OCRerr]UEST DOCUMENT NORMALIZED MEASURES [OCRerr]I[OCRerr]I;gp ??ORIGINALII "MODIFIED11 Request A9 33 1 (2 Relevant) 82 50 27 50 NR .5000 NR .8[OCRerr]38 NP .1718 NP e6780 Request B2 5 1 8o 17 2 (3 Relevant) 65 [OCRerr]2 `42 27 NR .7553 NR .83514 NP .14392 NP .7683 Request BlO 1 2 09 2 1 `48 19 3 (14 Relevant) 69 27 14 70 NR .8750 NR 1.0000 NP .7387 NP 1.0000 Request Bil 21 1 (1 Relevant) 36 NR .7531 NP 1.0000 NP .3091 NP 1.0000 Results of Four Requests in Which Manual Modification by [OCRerr]ortant Term Weighting is better than the Original Unmodified Search. Figure 18.