IRS13 Suffix Dictionaries 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. VI-5 INPUT AND EVALUATION STEM SUF[OCRerr]'IX `S' COLLECTION MATCHING FUNCTION MEASURE DICTIONARY DICTIONARY Abstract, Cosine Normede Recall .8954 .8817 mufleric Normed. Precision .6746 .6484 IRE-3 Abstract, Cosine Normed. Recall .8777 .8707 34 Logical Normed. Precision .6167 .6134 Requests Abstract, Overlap Normed. Recall .8725 .8408 Logical Normed. Precision .5829 .5611 CRAN-l 42 Abstract, Cosine Normed. Recall .8644 .8717 Requests Numeric Normed. Precision .6704 .7018 Text, Cosine Normed. Recall .7779 .7520 Numeric Normed. Precision .5573 .5308 Text, Cosine Normed. Recall .7695 .7768 Logical Normed. Precision .5248 .5462 Text, Overlap Normed. Recall .7434 .7546 ADI Logical Normed. Precision .4978 .5097 35 Requests Abstract, Cosine Normed. Recall .7601 .7253 N[OCRerr]eric Normed. Precision .5326 .4997 Abstract, Cosine Normed. Recall .7546 .7296 Logical Normed. Precision .5221 .5044 Title, Cosine Normed. Recall .6722 .6435 Numeric Normed. Precision .4537 .4209 Performance Results C[OCRerr]nparing Stem and Suffix `5' Dictionaries for Ten Options on Three Collections, using Normalized Recall and Precision. Figure 1.