Detection and elimination of duplicates from multidatabase searches.

A major problem in the review and synthesis of citations resulting from multidatabase searching is overlap in coverage among the databases. The ability to identify and eliminate duplicate references from a multidatabase search provides an important service and significant time savings for the end user. The Upjohn Company's Corporate Technical Library has developed software to detect and eliminate duplicates from literature searches captured in electronic format. Methods for the identification of duplicates and the merging and sorting of unique citations are discussed, together with the library's procedures for electronic data capture.