IRS13 Document Length 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. V-15 C0LLECTI0[OCRerr] DICTIONARY EVALUATION ABSTRACT TITLE MEASURE Stem Normed. Recall .8954 .8145 IRE-3 Normed. Precision .6746 .5547 3[OCRerr] Thesaurus-2 Normed. Recall .9191 .8436 Requests (Harris 2) Normede Precision .7072 .5945 Thesaurus-3 Normed. Recall .9268 .8430 (Harris 3) Normed. Precision .7382 .6068 CRAN-l Stem Normed. Recall . 864[OCRerr] .8112 [OCRerr]2 Normede Precision .6704 .6185 Requests Thesaurus-3 Normed. Recall .8833 .8374 (Harris) Normed. Precision .6955 .6420 Suffix `5' Normed. Recall .7253 .6435 Normed. Precision .4997 .4209 Stem Normed. Recall .7601 .6722 ADI Normed. Precision .5326 .4537 35 Thesaurus-i Normed. Recall .8016 .7324 Requests (Harris) Normed. Precision .6069 .5462 Thesaurus-2 Normed. Recall .7548 .6877 (Hastie) Normed. Precision .5190 .4649 Performance Results comparing abstracts with titles for nine dictionaries on three collections, using normalized recall and precision. Figure 5.