find_outmost_points function

Function which finds the outermost points in order to be used as initial solution in archetypal analysis

find_outmost_points(df, kappas)

Arguments

• df: The data frame with dimensions n x d
• kappas: The number of archetypes

Returns

A list with members:

1. outmost, the first kappas most frequent outermost points as rows of data frame
2. outmostall, all the outermost points that have been found as rows of data frame
3. outmostfrequency, a matrix with frequency and cumulative frequency for outermost rows

Warning

This is a rather naive way to find the outermost points of a data frame and it should be used with caution since for a n x d matrix we need in general 8 n^2/(2^30) GB RAM for numeric case. Check your machine and use it. As a rule of thumb we advice its usage for n less or equal than 20000.

Examples

data("wd2") #2D demo 
df = wd2
yy = find_outmost_points(df,kappas=3)
yy$outmost #the rows of 3 outmost points
yy$outmostall #all outmost found
yy$outmostfrequency #frequency table for all
df[yy$outmost,] #the 3 outmost points
#
###
#
data("wd3") #3D demo 
df = wd3
yy = find_outmost_points(df,kappas=4)
yy$outmost #the rows of 4 outmost points
yy$outmostall #all outmost found
yy$outmostfrequency #frequency table for all
df[yy$outmost,] #the 4 outmost points

See Also

find_furthestsum_points, find_outmost_convexhull_points,

find_outmost_projected_convexhull_points,

and find_outmost_partitioned_convexhull_points