Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

 Quantlet: xcfcgk Description: performs a fuzzy Gustafson-Kessel cluster analysis.

Reference(s):
James C. Bezdek (1981). Pattern Recognition with Fuzzy Objective Function Algorithm. Plenum Press, New York. James C. Bezdek and Sankar K. Pal (1992). Fuzzy Models for Pattern Recognition. IEEE Press, New York.

 Usage: {v,d,uu}=xcfcme(x,c,m,e) Input: x n x p matrix of n row points to be clustered c scalar representing the number of clusters m scalar, determines the fuzziness of clustering (m>1) e scalar, termination tolerance Output: v p x p matrix of cluster centers d n x p matrix of distances uu n x p matrix of result

Example:
```proc(xx,yy1,yy2)=idxgraph(out,x)
xx=matrix(max(x[,1])+1)*0
yy1=xx
yy2=xx
i=1
while(i <= max(x[,1])+1)
xx[i]= i-1
yy1[i]= max(paf(out.uu'[,1], x[,1]==i-1))
yy2[i]= max(paf(out.uu'[,2], x[,1]==i-1))
i=i+1
endo
endp
library("xclust")
library("plot")
x=z[,2:3]
c=2
m=2
e=0.01
out=xcfcgk(x,c,m,e)      //apply fuzzy gustafson-kessel clustering
out.uu'
tt=idxgraph(out,x)
tyy1=tt.xx~tt.yy1
tyy2=tt.xx~tt.yy2
tyy1 = setmask(tyy1, "line", "red", "thin")
tyy2 = setmask(tyy2, "line", "blue", "medium")
disp = createdisplay(1,1)
show(disp,1,1, tyy1, tyy2)
strm = string("%1.0f", m)
setgopt(disp,1,1, "title","Membership Function for Fuzzy GK with m="+ strm)

```
Result:
```gives the partitions of membership functions and depicts
this membership function for m = 2.
```

Author: H. Sofyan, 20020114
(C) MD*TECH Method and Data Technologies, 05.02.2006