272
Russell C. Coile
Professional Paper No. 189, July 1977, Error
Detection In Computerized Information Retrieval
Data Bases
Paper presented at the Sixth Cranfi eld International Conference
on Mechanized Information Storage and Retrieval Systems, Cranfi eld
Institute of Technology, Cranfi eld, Bedford, England, 26-29 July 1977
Abstract
The introduction of on-line interactive literature searching systems in
recent years has made it possible for information scientists to conduct
bibliometric studies which might have been diffi cult or impractical to do by
manual methods. The unconventional uses of on-line information retrieval
systems are becoming more common as we learn how to search using
non-subject information fi elds. Author's name, organizational affi liation,
journal's name, year of publication, etc., can now be searched for easily.
However, sometimes there are problems. For example, if the name of
the author in a database such as MEDLINE is given with initials for fi rst
and middle names, Bloggs, J. B. may be confused with Bloggs, J. B. since
Joseph Blackwell Bloggs may be a mathematician while James Blackwood
Bloggs is a chemist.
It would seem worthwhile for those responsible for management of
these mechanized information storage and retrieval databases to attempt
to use all economically feasible error-detecting and correcting schemes
to reduce the error rate as much as practicable. Several suggestions for
detecting errors have been examined.
Introduction
Mechanized information storage and retrieval systems have brought a
new era to information science and library operations. However, along with
added fl exibility and speed of searching and retrieval, we have become faced
with more stringent requirements for accuracy in databases. Unconventional
uses of on-line information retrieval systems are becoming increasingly
common as we learn how to search using non-subject information fi elds.