Techniques of Matching the Duplicate Items in the Bibliographic Data Bases

Abstract

Techniques of matching the duplicate items in the bibliographic data bases are greatly needed both for the compilation and maintenance of the data bases, and for the information retrieval processes. This paper aims to examine the formulars and use of the book identifying devices, the identifiers and descriptors of each document in the data bases, using two actual examples: (i) OCLC(The Ohio College Library Center) as an illustration of a on-line shared cataloging system, (ii) the project LOC(London i.e. The British Museum-Oxford-Cambridge) as an example of the identification of the books published before 1801, using the 'fingerprints' of the books. The author's concept of the 'quasi-identifier' is first introduced into this field.