findModelCluster {svcR} | R Documentation |
SvcR implements a clustering algorithm based on separator search in a feature space between points described in a data space. Data format is defined by an attribute/value table (matrix). The data are transformed within a kernel to a feature space into a unic cluster bounded with a ball radius and support vectors. We can used the radius of this ball in the data space to reconstruct the boundary shaped now in several clusters.
## S4 method for signature 'integer': findModelCluster(x=as.integer(1), MetLab=1, KernChoice=1, Nu=0.8, q=20, K=1, G=10, Cx=1, Cy=2, DName="iris", fileIn="") ## S4 method for signature 'character': findModelCluster.chargeMatrix( x="term", fileIn="c:/files") ## S4 method for signature 'matrix': findModelCluster.Eval( x=matrix() ) ## S4 method for signature 'numeric': findModelCluster.Test()
x |
means MetOpt parameter in standard use means DName in chargeMatrix use means DatMat in Eval use, a Matrix given as unic argument |
MetLab |
option taking value 1 (grid labelling) or 2 (mst labelling) or 3 (knn labelling) |
KernChoice |
option taking value 0 (Euclidian) or 1 (RBF) or 2 (Exponential) |
Nu |
kernel parameter |
q |
kernel parameter |
K |
number of neigbours on the grid |
G |
size of the grid |
Cx |
1st data coordinate to plot for 2D cluster extraction |
Cy |
2nd data coordinate to plot for 2D cluster extraction |
DName |
Name of data which is the prefix of files : ‘DName_mat.txt’, ‘DName_att.txt’, ‘DName_var.txt’ |
fileIn |
path where to find files as "D:R\library\svcR\" |
format of ‘DName_mat.txt’ (data matrix): 1 1 5.1 1 2 3.5 2 3 1.4 it mean mat[1, 1] = 5.1, mat[1, 2] = 3.5, mat[2, 3] = 1.4
format of ‘DName_att.txt’ : X1 X2 it mean X1 is the name of first column of the data matrix, X2 is the name of the second column of the data matrix
format of ‘DName_var.txt’ : v1 v2 it mean v1 is the name of first line of the data matrix, v2 is the name of the second line of the data matrix
An S4 object of class findModelCluster
The object is the svc model along with
the slots :
WVectorsYA |
lagrange coefficients : WVectorsYA$A |
Matrice |
variables names Matrice$var, attributes names Matrice$Att and data Matrice$Mat |
MatriceK |
kernel matrix |
Data |
Data Matrix |
MinMaxXY |
min max values for first and second coordinates |
MisClass |
missclassfied points |
DName |
prefix name of data for files decoding |
fileIn |
data file path |
ClassPoints |
class of grid points |
Cx |
x column id of data matrix |
Cy |
y column id of data matrix |
Nu |
nu value of the svc model |
KNN |
knn value for labelling |
SizeGrid |
size grid for labelling |
AroundNullVA |
almost null value for lagrange coefficient estimation |
NumPoints |
value fo grid points |
slots can be accessed by object@slot
Nicolas Turenne - INRA France nicolas.turenne@jouy.inra.fr
N.Turenne , Some Heuristics to speed-up Support Vector Clustering , technical report 2006, INRA, France http://migale.jouy.inra.fr/~turenne/svc.pdf
## exemple with iris data MetOpt = 1; # optimisation method with randomization MetLab = 1; # grid labelling KChoice = 1; # 0: eucli 1: radial 2: radial+dist Nu = 1.0; q = 2000; # lot of clusters K = 1; # only 1 nearest neighbour for clustering Cx = Cy = 0; # we use principal component analysis factors G = 13; # size of the grid for cluster labelling DName = "iris"; fileIn = ""; # fileIn migth be such as "D:/R/library/svc/", if NULL it will work on iris data fmc = findModelCluster( as.integer(MetOpt), MetLab, KChoice, Nu, q, K, G, Cx, Cy, DName, fileIn);