bpec {evclust} | R Documentation |
Belief Peak Evidential Clustering (BPEC)
Description
bpec
computes a credal partition from a matrix of attribute data using the
Belief Peak Evidential Clustering (BPEC) algorithm.
Usage
bpec(
x,
g,
type = "full",
pairs = NULL,
Omega = TRUE,
alpha = 1,
beta = 2,
delta = 10,
epsi = 0.001,
disp = TRUE,
distance = 1,
m0 = NULL
)
Arguments
x |
input matrix of size n x d, where n is the number of objects and d the number of attributes. |
g |
Matrix of size c x d of prototypes (the belief peaks). |
type |
Type of focal sets ("simple": empty set, singletons and Omega;
"full": all |
pairs |
Set of pairs to be included in the focal sets; if NULL, all pairs are included. Used only if type="pairs". |
Omega |
Logical. If TRUE (default), the whole frame is included (for types 'simple' and 'pairs'). |
alpha |
Exponent of the cardinality in the cost function. |
beta |
Exponent of masses in the cost function. |
delta |
Distance to the empty set. |
epsi |
Minimum amount of improvement. |
disp |
If TRUE (default), intermediate results are displayed. |
distance |
Type of distance use: 0=Euclidean, 1=Mahalanobis. |
m0 |
Initial credal partition. Should be a matrix with n rows and a number of columns equal to the number f of focal sets specified by 'type' and 'pairs'. |
Details
BPEC is identical to ECM, except that the prototypes are computed from delta-Bel graph using function
delta_Bel
. The ECM algorithm is then run keeping the prototypes fixed. The distance to the
prototypes can be the Euclidean disatnce or it can be an adaptive Mahalanobis distance as in the CECM
algorithm.
Value
The credal partition (an object of class "credpart"
).
Author(s)
Thierry Denoeux.
References
Z.-G. Su and T. Denoeux. BPEC: Belief-Peaks Evidential Clustering. IEEE Transactions on Fuzzy Systems, 27(1):111-123, 2019.
See Also
Examples
## Clustering of the Four-class dataset
## Not run:
data(fourclass)
x<-fourclass[,1:2]
y<-fourclass[,3]
DB<-delta_Bel(x,100,0.9)
plot(x,pch=".")
points(DB$g0,pch=3,col="red",cex=2)
clus<-bpec(x,DB$g0,type='pairs',delta=3,distance=1)
plot(clus,x,mfrow=c(2,2))
## End(Not run)