Performing dimension reduction with MDS
Multidimensional Scaling (MDS) is a technique to create a visual presentation of similarities or dissimilarities (distance) for a number of objects. The multi prefix indicates that one can create a presentation map in one, two, or more dimensions. However, we most often use MDS to present the distance between data points in one or two dimensions.
In MDS, you can either use a metric or a nonmetric solution. The main difference between the two solutions is that metric solutions try to reproduce the original metric, while nonmetric solutions assume that the ranks of the distance are known. In this recipe, we will illustrate how to perform MDS on the swiss dataset.
Getting ready
In this recipe, we will continue using the swiss dataset as our input data source.
How to do it...
Perform the following steps to perform multidimensional scaling using the metric method:
- First, you can perform metric MDS with a maximum of two dimensions:
> swiss.dist =dist...
