The Choice of Reference Poin ts in Best-Match File Searching

Improvements to the exhaustive search method
of best-match file searching have previously been 
achieved by doing a preprocessing step involving the
calculation of distances from a reference poin t. 
  This paper discusses the proper choice of reference
poin ts and extends the previous algorithm to use 
more than one reference poin t.  It is shown that reference
poin ts should be located outside of data clusters. 
 The results of computer simulations are presented which
show that large improvements can be achieved 
by the proper choice and location of multiple reference poin ts.

CACM May, 1977

Shapiro, M.

matching, file searching, best match, nearest-neighbor classification

3.73 3.74

CA770506 JB December 29, 1977  1:57 AM

2575	4	2966
2966	4	2966
3134	4	2966
2524	5	2966
2966	5	2966
2966	5	2966
2966	5	2966