simM {ipsfs}

IFS similarity measure simM

Description

IFS similarity measure values using simM computation technique with membership,non-membership, and hesitancy values of two objects or set of objects.

Usage

simM(ma, na, mb, nb, p, k)

Arguments

ma

IFS membership values for the data set x computed using either triangular or trapezoidal or guassian membership function

na

IFS non-membership values for the data set x computed using either Sugeno and Terano's or Yager's non-membership function

mb

IFS membership values for the data set y computed using either triangular or trapezoidal or guassian membership function

nb

IFS non-membership values for the data set y computed using either Sugeno and Terano's or Yager's non-membership function

p

Lp norm values for measuring the p-norm distance between x and y, values range from 1 to 5

k

A constant value, considered as 1

Value

The IFS similarity values of data set y with data set x

References

H. B. Mitchell. On the dengfeng–chuntian similarity measure and its application to pattern recognition. Pattern Recognition Letters, 24(16):3101 - 3104, 2003.

Examples

x<-matrix(c(12,9,14,11,21,16,15,24,20,17,14,11),nrow=4)
y<-matrix(c(11,21,6),nrow=1)
a<-mn(x)
b<-std(x)
a1<-mn(y)
b1<-std(y)
lam<-0.5
ma<-memG(a,b,x)
na<-nonmemS(ma,lam)
ha<-hmemIFS(ma,na)
mb<-memG(a1,b1,y)
nb<-nonmemS(mb,lam)
hb<-hmemIFS(mb,nb)
p<-2
k<-1
simM(ma,na,mb,nb,p,k)
#[1] 0.3840287 0.3837673 0.3849959 0.3849959