 |
 |
|
If you are interested in information about our bibliographic services, please call or email us:
PRODUCT & SERVICE INFORMATION
|
|
|
|
|
Phone: (800) 474-9816
Email: support@brodart.com
|
|
|
|
|
If you are uncertain where to find help at Brodart, start by contacting our customer service department:
CUSTOMER SERVICE
|
|
|
|
|
Phone: (800) 474-9802
Email: bookscs@brodart.com
|
|
|
|
|
|
Consolidation
Duplicate records of the same title often appear on a bibliographic file. This is sometimes caused by libraries merging their local databases or by a single library importing records into their local system. Consolidation reduces the number of occurences of the "same title." It does so by comparing each record to all the others in the database and then by following rules for defining "duplicates" as duplicates aren't always identical. Rules must also be established to determine which "matching" record is retained and which removed from a database.
Brodart offers a number of standard options for identifying duplicate records and then consolidating them:
Consolidation Methods
|
 |
LCCN - This option uses the LCCN (010$a) plus any prefixes.
|
|
|
ISBN - This option uses the ten-character ISBN (020$a) excluding punctuation and spaces.
|
Advanced Consolidation Methods
A consolidation can be designed which is based on matching a combination of bibliographic data points. "Match keys" are created for each record and compared to all other "match keys" in a database. For records to match (so that duplicates can be deleted) the records must match on ALL the standard data points defined below, but many other optional points are available.
|
|
The first 15 characters of the title (MARC tag 245, $a and $b) and the last ten characters of the title, ignoring leading articles "A," "An" and "The" and ignoring punctuation. This data must match exactly for records to consolidate.
Example source data:
($a) American history, American television:
($b) interpreting the video past
Resulting Key:
AMERICAN HISTORVIDEO PAST
(In this example, the presence or absence of a subfield delimiter b is insignificant.)
|
|
|
The first five characters of the GMD (245$h). This data must match exactly for records to consolidate.
|
|
|
The first four Arabic numerals of the Imprint (260$c), bypassing any leading date preceded by a "c." If no data is found, a match is assumed.
|
|
|
The first numeric sequence in the physical description (300$a), plus or minus 10. If no 300 field is found, the processing matches to a 301 tag, plus or minus 1. (A 301 tag will not match to a 300 tag.) If no data is found, a match is assumed.
|
Examples of optional match points are publisher (260$b) and edition (250$a).
Master Record Retention
For any of the three consolidation options a hierarchy can be specified to establish which of the "matching" records should become the master record retained on the file. This heirarchy can be by library, by cataloging source, by preference given to records with OCLC numbers, etc.
|
|
|
|
