[Zebralist] Phrase searches that cross subfield bounds

Giannis Kosmas kosmas at lib.uoc.gr
Mon Aug 29 15:41:55 CEST 2005


I tried to use this indexing mechanism and I want to comment the following.

1. The underscore as prefix and suffix, is ignored in the resulting term.
For example the subject:
(3,650)
    (3, 0)
        (3,a) Political science
        (3,x) Dictionaries

with index formula : elm mc-650.._$a__$x__$y__$z_ 

produces: (3,mc-650.._$a__$x__$y__$z_) Political scienceDictionaries

2. A subfield's position in the index formula in regard to the 
subfield's position in the MARC field is significant for the indexing 
process.

For example, given these two subject headings,

(3,650)
    (3, 0)
        (3,a) Natural law
        (3,x) Study and teaching
        (3,z) Spain.

(3,650)
    (3, 0)
        (3,a) Political science
        (3,z) Spain
        (3,x) History.

The index formula "elm  mc-650.._$a__$x__$y__$z_" will produce the terms:
(3,mc-650.._$a__$x__$y__$z_) Natural law Study and teaching Spain.
and
(3,mc-650.._$a__$x__$y__$z_) Political science History.
but not the term:
(3,mc-650.._$a__$x__$y__$z_) Political science Spain History.

Is there any way to specify that the order of a set of subfields as they 
appear in a MARC record is not significant?
If not, I think that it would be a nice feature.

Thanks in advance,
Giannis

-- 
Giannis Kosmas                    <kosmas (at) lib.uoc.gr>

University of Crete                Tel : +30 2810 393254
Library Automation Team            Fax : +30 2810 393250
         Knossou Ave. 71409, Heraklio, GREECE







More information about the Zebralist mailing list