Multidimensional Scaling and Genetic Algorithms: A Solution Approach to Avoid Local Minima
- Multidimensional scaling is very common in exploratory data analysis. It is mainly used to represent sets of objects with respect to their proximities in a low dimensional Euclidean space. Widely used optimization algorithms try to improve the representation via shifting its coordinates in direction of the negative gradient of a corresponding fit function. Depending on the initial configuration, the chosen algorithm and its parameter settings there is a possibility for the algorithm to terminate in a local minimum. This article describes the combination of an evolutionary model with a non-metric gradient solution method to avoid this problem. Furthermore a simulation study compares the results of the evolutionary approach with one classic solution method.
| Author: | Stefan Etschberger, Andreas HilbertGND |
|---|---|
| URN: | urn:nbn:de:bvb:384-opus4-2371 |
| Frontdoor URL | https://opus.bibliothek.uni-augsburg.de/opus4/292 |
| Series (Serial Number): | Arbeitspapiere zur Mathematischen Wirtschaftsforschung (181) |
| Publisher: | Institut für Statistik und Mathematische Wirtschaftstheorie, Universität Augsburg |
| Place of publication: | Augsburg |
| Type: | Working Paper |
| Language: | English |
| Date of Publication (online): | 2006/07/31 |
| Year of first Publication: | 2002 |
| Publishing Institution: | Universität Augsburg |
| Release Date: | 2006/07/31 |
| GND-Keyword: | Multidimensionale Skalierung; Genetischer Algorithmus |
| Page Number: | 21 |
| Institutes: | Wirtschaftswissenschaftliche Fakultät |
| Wirtschaftswissenschaftliche Fakultät / Institut für Statistik und mathematische Wirtschaftstheorie | |
| Dewey Decimal Classification: | 3 Sozialwissenschaften / 31 Statistiken / 310 Sammlungen allgemeiner Statistiken |
| Licence (German): |  Deutsches Urheberrecht |
