University of Konstanz
Graduiertenkolleg / PhD Program
Computer and Information Science

Graduation Talks


Engineering Multidimensional Scaling Approaches to Graph Drawing


Mirza Klimenta, University Konstanz
Konstanz, Germany

date & place

Wednesday, 29.06.2011, 16:15 h
Room C 252


Effective graph drawings are often a result of specific energy model optimization, an objective common to other fields. Recently it has been shown that Multidimensional Scaling (MDS) techniques tailored for data analysis about proximity can be successfully exploited. The approaches of Classical Scaling (CMDS) and Distance Scaling (DS) usually satisfy the basic aestetic criteria on the layout. Due to more direct distance fitting and allowing for control over different distance influence, DS is the preferred choice (see Fig. 1). While with complexity requirements prohibitive for large graph drawing, the basic methods o er a promising directionfor various visualization purposes.
The objective of this thesis is to refine and improve the use of MDS for Graph Drawing (GD) in terms of scalability and flexibility.