Dissertation Information for Leny Struminger NAME: - Leny Struminger
DEGREE:
- Ph.D.
DISCIPLINE:
- Library and Information Science
SCHOOL:
- Rutgers University (USA) (1984)
ADVISORS: - Henry Voos
COMMITTEE MEMBERS: - Thomas H. Mott Jr. - Hendrik Edelman - Charles Meadow
MPACT Status: Fully Complete
Title: METHODOLOGY FOR EVALUATING RELATED DATABASES AND A COMPUTERIZED ALGORITHM FOR MERGING THEIR OUTPUT
Abstract: The last decade witnessed a proliferation of computerized databases in many areas. With the development of large computer storage capacity and extensive low cost data communications systems access to computerized databases increased tremendously. The latest developments in microcomputers and the availability of personal computers to masses of users added to the requirements of accessing databases and further influenced their developments. Today, there are many databases addressing similar topics, and it is a more commonly accepted procedure to search more than one database in order to retrieve pertinent information to a query.
Databases are produced by different organizations and there is not at present a standardized record format. Organizing the retrieved documents from each database in a single, useful list of citations has become an important issue in database searching.
Two related bibliographic databases in the field of medicine (Excerpta Medica and Medline) have been analyzed for "file content" and "record content" and observations gathered on the treatment of journal articles in each of the databases. A sample of 331 articles indexed by both databases from 22 randomly selected journals in the field have been used in order to study the consistency in the representation of the original documents.
From the patterns observed in the treatment of titles, authors, abstracts, volume, issue and pagination by year of publication and by journal, recommendations for record standardization are made.
The pagination field was found to be consistently correct and in agreement in 99% of the cases studied and the author field after editing was also found to be in agreement in 97% of the cases.
Based on these findings a programmable computer algorithm was developed to identify duplicate citations among related databases. The methods used in developing the comparison algorithm can be applied to other computerized databases. The resulting programs can be stored in microcomputers and used as a useful tool for editing the output from cross-database searching by providing the user with a more practical bibliography.
| |
MPACT Scores for Leny Struminger A = 0
C = 1
A+C = 1
T = 0
G = 0
W = 0
TD = 0
TA = 0
calculated 2008-01-31 06:09:38
Advisors and Advisees Graph
generating graph, please reload |