3-point pseudo-scale distance measure for measuring indirect proximity
A typical proximity measure employed to find the distance between two points in multidimensional scaling techniques (MDS) is the Euclidean distance. This direct measurement approach however lacks the ability to extract more information from other points in the data. In this paper, we introduce a 3-p...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Published: |
Springer New York LLC
2013
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/50461/ https://www.researchgate.net/publication/288742329 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | A typical proximity measure employed to find the distance between two points in multidimensional scaling techniques (MDS) is the Euclidean distance. This direct measurement approach however lacks the ability to extract more information from other points in the data. In this paper, we introduce a 3-point pseudo-scale distance measure based on Kendall's tertiary treatment of ties (TTT). This distance is coupled with ANOVA-Tukey's test approach to reveal the hidden structure where the Euclidean distance used in multidimensional scaling technique is not able to cater. To demonstrate the proposed technique, a set of rainfall data recorded between the years 1968 and 2003 over nine rain-gauge stations in Peninsular Malaysia are used |
|---|