Authormagic - Automated author disambiguation and identification of individual researchersIn the field of information science and library science, the
problem of matching names, as they appear on a document, to
an individual researcher gains of more and more importance.
A few decades ago, this particular problem was not a high-
priority issue, hence the quantity of documents produced had
been lower, and the researchers had known each other across
the respective field. Nowadays, due to the immense growth of
the scientific community with ever-new generations, this
level of familiarity is no longer existent. The scientific
throughput in terms of the mere quantity of documents
produced by researchers is steadily growing with emerging
technologies and research sources. Other reasons for the
increased number of scientific publications might be the
stringent interest of the general public in science and
elevated numbers of contributions to the scientific
communication from rising industrial nations (e.g. China or
India). Despite the effort spent in trying to keep an
overview over the authors, many identities behind author
names remain unknown.The goal of this project is to recreate the match between
name and individual. Since the final count of individuals is
unknown beforehand, a grouping by last-name is performed as
a first step. Within these groups, all documents have to be
identified that had been written or co-authored by members
of the particular group. Information from these documents
can be used for identifying the individual--examples for
this information are the name of the author, the affiliation
of the author, the date on the paper, keywords used to tag
the paper, (self-) citations, co-authorship and others. With
the help of these characteristics, clusters can be formed
within the groups which will ideally represent the profile
of an individual researcher. With sufficient data it is
hence possible to determine if the name written on the
document (e.g. “Ellis, J.”) is one real person or another
(e.g. “Ellis, John” or “Ellis, Jordan”). The knowledge about
the tangible authorship allows the development of entirely
new services. One example for such a service would be the
automated and objective ranking of authors based on their
scientific quality to substitute conventional methods of
quality assessment used today.
| Project manager: Prof. i. R. Dr. Klaus Meyer-Wegener, Dr. Mele, Salvatore
Project participants: Dr.-Ing. Henning Weiler
Duration: 1.2.2009 - 16.2.2012
Sponsored by: CERN Wolfgang-Gentner-Stiftung
Contact: Meyer-Wegener, Klaus Phone +49.9131.85.27892, E-Mail: Klaus.Meyer-Wegener@fau.de
| Publications |
---|
Weiler, Henning ; Meyer-Wegener, Klaus ; Mele, Salvatore: Authormagic - An Approach to Author Disambiguation in Large-Scale Digital Libraries. In: Macdonald, Craig ; Ounis, Iadh ; Ruthven, Ian (Ed.) : Proc. 20th ACM Conf. on Information and Knowledge Management (CIKM 2011 Glasgow, Scotland, UK October 24-28). New York : ACM, 2011, pp 2293-2296. - ISBN 978-1-4503-0717-8 |
|