Trends in Astronomical Publication
and
Information Retrieval Systems
Why is it important?
The end result of most astronomical research is one or more publications
in refereed astronomy or astrophysics journals, so it behooves astronomers
to understand how and why we publish. Understanding the phenomenon of astronomical
publication can better enable astronomers to evaluate candidates for jobs
and grants and can help decide questions about the future of astronomical
journals.
In addition to our research findings about astronomical publication, we
developed sophisticated utilities to improve data consistency and to extract
topic information from documents in databases using state-of-the-art text
categorization. Astronomy is uniquely suited for this work because abstracts
from a very large fraction of astronomical papers are available electronically.
The browse facilities developed in this research will also be usable by researchers
in other fields whose electronic databases are not so complete. Digital libraries
are becoming increasingly important in science and in society as a whole,
and effective access to this information hinges on consistent, high quality
data and a wide variety of data retrieval tools that can efficiently access
this data.
``The investigator is staggered by the findings and conclusions
of thousands of other workers-conclusions which he cannot find the time to grasp, much
less remember, as they appear.''
--Vannevar Bush, 1945
What did we do?
I organized an interdisciplinary research group to use the NASA
Astrophysics Data System
database of astronomical abstracts to study a large number of trends in
astronomical publication.
Allison Powell
and Jim French
, members of the UVa Department of Computer
Science
, developed tools to analyze the database.
Guenther Eichhorn
, Michael Kurtz
, and Steve Murray, of the
Smithsonian Astrophysical Observatory
, supplied the database and helped identify the various selection effects
that we had to take into account. I posed the questions we would like to
answer and analyzed the data that Allison and Jim extracted from the database,
organized the research effort, and wrote up the results.
On the database management side of things, Allison and Jim developed
innovative techniques to extract information from databases in general and
the ADS database in particular, and I acted as the ``domain expert,'' deciding
what information is useful and making sure that the utilities were correctly
extracting the necessary data.
"Automating the Construction of Authority Files in Digital Libraries:
A Case Study," James C. French
, Allison L. Powell
, Eric Schulman
, and John L. Pfaltz
, 1997, in Research and Advanced Technology for Digital Libraries, ed. C.
Peters and C. Thanos, Springer-Verlag Lecture Notes in Computer Science, 1324,
55 (refereed conference proceedings).