IRS13 Evaluation Parameters 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. [OCRerr]L3 Number of Documents Examined 111111 III Best Possible Precision Performance Curve 1.0 Precision 1.0 - 0.8 0.8 0.6 0.6 - 0.4 - 0.4 x[OCRerr]l5 0.2 - I I [OCRerr]loc I I [OCRerr] 0 0.2 0.4 0.6 0.8 2[OCRerr]0 Recall l.0~ 0.2 - N ½>[OCRerr][OCRerr]1 I I 1.0 [OCRerr] Rec( 0 0.2 0.4 0.6 0.8 a) Document Cut-Off "Pseudo-Cr[OCRerr]field" c[OCRerr]puted by hand. b) Document Cut-off - "Pseudo-Cranfield as c[OCRerr]nputed by evaluation routines a ten recall levels. Cran-l, Abstracts, Thesaurus-3, Micro Evaluation of 142 Requests An Illustration of the Document Cut-off or "Pseudo-Cranfield Method. Fig. 114