map.summary {popsom7} | R Documentation |
Summary Object for Map
Description
Generate a summary object for 'map' objects.
Usage
map.summary(map,verb=TRUE)
Arguments
map |
An object of type 'map'. |
verb |
A switch controlling the output. |
Value
An object of type 'summary.map' which contains two structures:
- training.parameters
A dataframe containing the parameters the map was trained with.
- quality.assessments
A dataframe containing the quality assessments of the map. In particular, it contains the 'convergence' of the map which is a linear combination of variance capture and topographic fidelity of the map. A value close to 1 means a converged map (for more details see the reference below). Furthermore, it contains the 'separation' of the clusters. This is computed by the formula,
1 - wcss/bcss
In general, a value close to 1 means well separated clusters.
If 'verb' is TRUE the summar.map object will be formatted and printed to the screen, otherwise it will be returned as a data structure.
Author(s)
Lutz Hamel
References
Self-Organizing Map Convergence, Robert Tatoian and Lutz Hamel. Proceedings of the 2016 International Conference on Data Mining (DMIN'16), pp92-98, July 25-28, 2016, Las Vegas, Nevada, USA, ISBN: 1-60132-431-6, CSREA Press.
Examples
data(iris)
## set data frame and labels
df <- subset(iris,select=-Species)
labels <- subset(iris,select=Species)
## build a map
m <- map.build(df,labels,xdim=15,ydim=10,train=10000)
## compute a summary object and display it
s <- map.summary(m,verb=FALSE)
s