credal_RI {evclust}

Credal Rand indices

Description

credal_RI computes generalizations of the Rand index to compare credal partitions, as defined in Denoeux et al (2018).

Usage

credal_RI(P1, P2, type = "c")

Arguments

P1

Relational representation of the first credal partition such as generated by function pairwise_mass

P2

Relational representation of the second credal partition such as generated by function pairwise_mass

type

"c" for degree of conflict (default), "j" for Jousselme's distance and "b" for belief distance.

Details

In Denoeux et al. (2018), two generalizations of the Rand index for comparing credal partitions are defined: one is based on distances between mass function, the other one is based on distances. In the latter case, two distances are proposed: Jousselme's distance and the L1 distance between belief functions. These three indices can be computed by function credal_RI.

Value

The credal Rand index

References

T. Denoeux, S. Li and S. Sriboonchitta. Evaluating and Comparing Soft Partitions: an Approach Based on Dempster-Shafer Theory. IEEE Transactions on Fuzzy Systems, 26(3):1231-1244, 2018.

See Also

nonspecificity, pairwise_mass

Examples

## Butterfly data
data(butterfly)
clus1<-kevclus(butterfly,c=2) 
P1<-pairwise_mass(clus1)
clus2<-ecm(butterfly,c=2)
P2<-pairwise_mass(clus2)
RI1<-credal_RI(P1,P2,"c")
RI2<-credal_RI(P1,P2,"j")
RI3<-credal_RI(P1,P2,"b")
print(c(RI1,RI2,RI3))