Scoring the Quality of E-Serials MARC Records Using Java

ABSTRACT

In 2015, the University of Illinois at Urbana–Champaign (UIUC) adopted a policy of cataloging electronic serials in the online catalog to represent the Library's large electronic journal holdings. Historically there had been no consistent policy of cataloging e-serials at UIUC, so this decision also required retroactive cataloging of the Library's existing holdings. In order to add the more than 100,000 titles to which the library has access, batch searching and bulk loading records are necessary. This can be particularly challenging with serials records, due to bibliographic changes (such as title and publisher) and record duplication in OCLC WorldCat. This column describes a Java program, written by the author, which aids in selecting high-quality records from batch searches to be bulk loaded into the catalog. The program scores the quality of Machine-Readable Cataloging Record (MARC) records based on data in a select number of fields and discards records that fail to meet minimum criteria, such as correct format or language of cataloging. The program ultimately ensures that certain base criteria are met, while creating more efficiency in the process of bulk loading e-serials records.