elbow {Kira}

Elbow method to determine the optimal number of clusters.

Description

Generates the Elbow graph and returns the ideal number of clusters.

Usage

elbow(data, k.max = 10, method = "AutoElbow", plot = TRUE, 
      cut = TRUE, title = NA, xlabel = NA, ylabel = NA, size = 1.1,  
      grid = TRUE, color = TRUE, savptc = FALSE, width = 3236, 
      height = 2000, res = 300, casc = TRUE)

Arguments

data

Data with x and y coordinates.

k.max

Maximum number of clusters for comparison (default = 10).

method

Method used to find the ideal number k of clusters: "jump", "curvature", "Exp", "AutoElbow" (default).

plot

Indicates whether to plot the elbow graph (default = TRUE).

cut

Indicates whether to plot the best cluster indicative line (default = TRUE).

title

Title of the graphic, if not set, assumes the default text.

xlabel

Names the X axis, if not set, assumes the default text.

ylabel

Names the Y axis, if not set, assumes the default text.

size

Size of points on the graph and line thickness (default = 1.1).

grid

Put grid on graph (default = TRUE).

color

Colored graphic (default = TRUE).

savptc

Saves the graph image to a file (default = FALSE).

width

Graphic image width when savptc = TRUE (defaul = 3236).

height

Graphic image height when savptc = TRUE (default = 2000).

res

Nominal resolution in ppi of the graphic image when savptc = TRUE (default = 300).

casc

Cascade effect in the presentation of the graphic (default = TRUE).

Value

k.ideal

Ideal number of clusters.

Author(s)

Paulo Cesar Ossani

References

Erich, S. Stop using the Elbow criterion for k-means and how to choose the number of clusters instead. ACM SIGKDD Explorations Newsletter. 25 (1): 36-42. arXiv:2212.12189. 2023. doi: 10.1145/3606274.3606278

Sugar, C. A. and James, G. M. Finding the number of clusters in a dataset: An information-theoretic approach. Journal of the American Statistical Association, 98, 463, 750-763. 2003. doi: 10.1198/016214503000000666

Zhang, Y.; Mandziuk, J.; Quek, H. C. and Goh, W. Curvature-based method for determining the number of clusters. Inf. Sci. 415, 414-428, 2017. doi: 10.1016/j.ins.2017.05.024

Onumanyi, A. J.; Molokomme, D. N.; Isaac, S. J. and Abu-Mahfouz, A. M. Autoelbow: An automatic elbow detection method for estimating the number of clusters in a dataset. Applied Sciences 12, 15. 2022. doi: 10.3390/app12157515

Examples

data(iris) # data set

res <- elbow(data = iris[,1:4], k.max = 20, method = "AutoElbow", cut = TRUE, 
             plot = TRUE, title = NA, xlabel = NA, ylabel = NA, size = 1.1, 
             grid = TRUE, savptc = FALSE, width = 3236, color = TRUE, 
             height = 2000, res = 300, casc = FALSE)
             
res$k.ideal # number of clusters