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-[OCRerr]9 CASE REQUEST DATA RETRIEVAL CHARACTERISTICS ABSTRACT TITLE Request [OCRerr]66, Number of matching concepts 3 1 6 Request Concepts, Sum of doct. wts. of match- 7 1 A Relevant Document mg concepts 965 Cosine numeric correlation .2619 .1666 Rank position 9 27 Request Ql22, Number of matching concepts 3 2 9 Request Concepts, Sum of doct. wts. of match- 6 2 B Relevant Document mg concepts 713 Cosine numeric correlation .1383 .1571 Rank positi[OCRerr]n [OCRerr]7 25 Request Q250, Number of [OCRerr]atching [OCRerr]oncepts 1 1 [OCRerr] Request Concepts, Sum of doct. wts. of match- 3 1 C mg concepts Relevant Document Cosine numeric correlation .2022 .1212 36G Rank position 3 15 Request Ql16, Number of matching concepts 3 3 13 Request Con- [OCRerr]um of doct. wts. of match- 7 3 D cepts, mg concepts Relevant Document Cosine numeric correlation .1277 .2631 57[OCRerr] Rank position 27 5 Four cases of request/relevant document analysis [OCRerr]omparing abstract[OCRerr] and titles, stem dictionary, Cran-l Collection. Figure 3[OCRerr].