GIC {GIC}

A General Iterative Clustering Algorithm

Description

An algorithm improves the proximity matrix (PM) from a random forest (RF) and the resulting clusters from an arbitrary cluster algorithm, such as PAM, as measured by the silhouette_score. The first PM that uses unlabeled data is produced by one of many ways to provide psuedo labels for a RF. After running a cluster program on the resulting initial PM, cluster labels are obtained. These are used as labels with the same feature data to grow a new RF yielding an updated proximity matrix. This is entered into the clustering program and the process is repeated until convergence.

Usage

GIC(data,cluster,initial="breiman",ntree=500,
        label=sample(1:cluster,nrow(data),replace = TRUE))

Arguments

data

an input dataframe without label

cluster

The number of clusters in the solution

initial

A method to calculate initial cluters to begin the iteration (default breiman). breiman: using Breimans' unsupervised method to find initial cluters, or purpose: using Siegel and her colleagues' purposeful clustering method to find initial cluters

ntree

the number of trees (default 500).

label

A truth set of labels, only required if purpose is used as the method to find the initial PM

Details

This code include Breimans' unsupervised method and Siegel and her colleagues' purposeful clustering method to calculate initial labels To imput user specified initial labels, please use the function initial

Value

An object of class GIC, which is a list with the following components:

PAM

output final PAM information

randomforest

output final randomforest information

clustering

A vector of integers indicating the cluster to which each point is allocated.

silhouette_score

A value of mean silhouette score for clusters

plot

A scatter plot which X-axis, y-axis, and color are first important feature, second important feature, and final clusters, respectively.

References

Breiman, L. (2001), Random Forests, Machine Learning 45(1), 5-32.

Siegel, C.E., Laska, E.M., Lin, Z., Xu, M., Abu-Amara, D., Jeffers, M.K., Qian, M., Milton, N., Flory, J.D., Hammamieh, R. and Daigle, B.J., (2021). Utilization of machine learning for identifying symptom severity military-related PTSD subtypes and their biological correlates. Translational psychiatry, 11(1), pp.1-12.

Examples

data(iris)
##Using breiman's method
rs=GIC(iris[,1:4],3,ntree=100)
print(rs$clustering)