IRS13 Search Matching Functions 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. III- 31 CoL![OCRerr]CTION INPUT and DICTIONARY EVALUATION LOGICAL NUMERIC MEASURE VECTORS VECTORS Abstract, Suffix `S' Normed Recall .8707 .8817 Normed Precision .6l3[OCRerr] .648[OCRerr] IRE-3 Abstract, Stem Normed Recall .8777 .895[OCRerr] 3[OCRerr] Normed Precision .6167 .67[OCRerr]6 Requests Abstract, Thesaurus Normed Recall .9067 .9268 -3 Normed Precision .6574 .7382 Abstract, Stem Normed Recall .8397 .8644 Normed Precision .6377 .6704 CRAN-l Abstract, Thesaurus Normed Recall .8729 .8837 42 -3 Normed Precision .6936 .6952 Requests Title, Stem Normed Recall .8120 .8112 Normed Precision .6212 .6185 Text, Suffix `5' Normed Recall .7768 .7520 Normed Precision .5462 .5308 Text, Stem Normed Recall .7695 .7779 Normed Precision .5248 .5573 ADI Text, Thesaurus-i Normed Recall .7819 .8206 Normed Precision .5092 .6273 35 Requests Text, Thesaurus-2 Normed Recall .6589 e7774 (Hastie) Normed Precision .3602 .5441 Abstract, Suffix `5' Normed Recall .7296 .7253 Normed Precision .5044 .4997 Abstract, Stem Normed Recall .7423 .7601 Normed Precision .4904 .5326 Abstract, Thesaurus Normed Recall .8043 .8016 -1 Normed Precision .5823 .6069 Performance results c[OCRerr]xiparing numeric and logical vectors using cosine corre- lation for 13 options on three coliections, using normalized recall and precision. Figure 17.