IRS13 Test Environment 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. Average Recall at Fallout Average Fallout Precision PrE Rank of Given Versus Rank of at Given at Given VE Last Document Recall Normalized Normalized First & Document Document RE ~n Relevant Cut-off Curve, Recall Precision Second Cut-off Cut-off C[OCRerr] (Fig. 17) (e.g. Fig Pseudo Relevant (e.g. Fig. (e.g. Fig. Ps 16) Cut-off (Fig. 18) 16) 16) Cu S S S S S G G G S S S G S G G G S S S S S G G G S S S S S G G G S Specific Requests Perform best G General Requests Perform best All results are based on Abstracts with Stem Dictionary Summary of Results of Nine Ways of Comparing Specific and General Requests Fig. 15