TITLE: What should I research next? Identifying influences, ideas and opportunities by data mining citations using formal concept analysis.

PRESENTER: Prof. Peter W. Eklund, http://www.kvocentral.org, http://www.mail-sleuth.com

AFFILIATION: Faculty of Informatics University of Wollongong

DATE: Friday 9th July 2004

TIME: 12:00noon - 1.00pm

PLACE: CSE K17 1st Floor Seminar Room

ABSTRACT:

One of the side effects of scientific work is the obligatory analysis of the field of research being studied. In my talk, FCA is presented as a means to analyzing a field of research using citations as input. In particular, results are presented based on a case study of 47 academic papers in a field of research: namely the application of formal concept analysis for software engineering. Inputs for the analysis are derived from the on-line ResearchIndex - formerly known as "CiteSeer". The analysis identifies inferences about the field of study based on domain background knowledge derived from the ISO12207 software engineering standard. General inferences about the diffusion of ideas are also presented based on author collaboration and citation counts. FCA reveals useful insights about the nature of the subject matter as well as the scientific community under examination.

BIOGRAPHY OF SPEAKER:

Peter W. Eklund is a Professorial Fellow in the Faculty of Informatics at the University of Wollongong. During the last 5 years Peter has been involved in the application of concept lattices to information retrieval and software re engineering. In late 2003, after several years development, a commercial version of the Conceptual Email Manager (called Mail-Sleuth), was released by Email Analysis Pty Ltd.

Host:

Mike Bain, Senior Lecturer
Tel: +61-2-9385-6935
Fax: +61-2-9385-4936
Email: mike@cse.unsw.edu.au
School of Computer Science & Engineering, UNSW.

Seminar Convenor:

Bernhard Hengst
Tel: +61-2-9385-3988,
E-mail: bernhardh@cse.unsw.edu.au
School of Computer Science & Engineering, UNSW.