ISR10 The Query-Document Matching Function chapter Joseph John Rocchio Harvard University Gerard Salton Use, reproduction, or publication, in whole or in part, is permitted for any purpose of the United States Government. 4-4 Keywords a b C d e f documents d1 d 2 d3 d 4 1 1 1 1 1 1 1 1 1 1 1 1 1 [OCRerr]ery 1 1 1 1 1 a) Query and document keyword images represented by a binary occurrence matrix. d1 [OCRerr] d2 D q R = [OCRerr]dI[OCRerr] : d[OCRerr] [OCRerr] *d3 D q R [OCRerr]d2, [OCRerr] d4 [OCRerr] b) Retrieval by set inclusion matbhing. L n([OCRerr]'(\ d) n(q [OCRerr] d)/n(q [OCRerr] d1 2 d1 2/5 d2 d2. 3/5 d4 1 d4 1/5 c) Relevance. values assi[OCRerr]ned by overlap correlation. Set Image Matcbing Operations Fignre 4.1 d) Relevance values assigned by metric corr. (1-metric d[OCRerr]s[OCRerr].)