ISR10
Scientific Report No. ISR-10 Information Storage and Retrieval
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.
4-1[OCRerr]
R = : d. [OCRerr]
1
Consider two document ima[OCRerr]es such that:
d [OCRerr]q and
£or arbitrary q. Then i£d1 : [OCRerr] d2 either:
d1 [OCRerr] d2 ; or d2[OCRerr].d1 ; or di [OCRerr] d2 and d2[OCRerr]d1
The equality case has already been considered; an [OCRerr]:.O£ the;other.three
possibilities lead to the existence 0£. some query which will retrieve
one but not the other 0£ the documents dl and d2. Thus in [OCRerr]eneral,
the relation [OCRerr] induced by set inclusion matbhin[OCRerr] is not an equivalence
relation.
In the case 0£ metric distance matchin[OCRerr] (o£ set or vector
represented indeximages), de£ine the retrieved set R by the condition:
:[OCRerr](q,d.)<[OCRerr]0
[OCRerr]ow consider two documents, d1 and d2, such that:
[OCRerr](d1,d2) =
where [OCRerr] is the smallest distance possible in terms 0£ the quantization