figen: Initialization of membership degrees over class range of a...

In inaparc: Initialization Algorithms for Partitioning Cluster Analysis

Initialization of membership degrees over class range of a selected feature

Description

Initializes the membership degrees matrix by using the class range of the coefficient of variation of a selected feature in the data set being processed.

Usage

figen(x, k, mtype, sfidx)

Arguments

x	an data.frame or matrix for the data set.
k	an integer for the number of clusters.
mtype	a character representing the type of membership degrees to be generated. The default type is f for generating fuzzy membership matrix. Use h for creating an hard (crisp) membership matrix.
sfidx	an integer for the column index of a selected feature. The default is the column index of a feature whose coefficient of variation is the maximum among all features in the data set.

Details

The function figen generates a numeric matrix containing the fuzzy initial membership degrees.

Value

an object of class ‘inaparc’, which is a list consists of the following items:

u	a numeric matrix containing the initial membership degrees.
sfidx	an integer for the column index of the selected feature, which used for random sampling.
call	a string containing the matched function call that generates this ‘inaparc’ object.

Author(s)

Zeynel Cebeci, Cagatay Cebeci

References

Cebeci, Z. (2018), "Initialization of Membership Degree Matrix for Fast Convergence of Fuzzy C-Means Clustering", In Proc. of 2018 International Conference on Artificial Intelligence and Data Processing (IDAP), IEEE, Sep. 2018, pp. 1-5., doi: 10.1109/IDAP.2018.8620920

See Also

imembrand, imembones

Examples

data(iris)

# Generate a fuzzy membership matrix using the 1st feature
u <- figen(iris[,1:4], k=5, sfidx=1)$u
head(u)
tail(u)

# Generate a fuzzy membership matrix using the internally determined feature
res <- figen(iris[,1:4], k=5)
u <- res$u
head(u)
tail(u)

inaparc documentation built on June 16, 2022, 5:09 p.m.