Power Fuzzy Clustering and Cluster-Wise Regression

Implementations of Power Fuzzy Clustering (PFC) and Power Fuzzy Cluster-wise Regression (PFCR) for multivariate data. The package supports Minkowski distances, with the L1 case solved via iteratively re-weighted least squares and the case p > 1 solved via coordinate-wise root finding, as well as an adaptive, regularised Mahalanobis distance with per-cluster covariance matrices. Both plain fuzzy clustering and cluster-wise linear regression are provided. The corresponding paper can be found at Nguyen P.T., Tortora C., and Punzo A. (2026) .

pfclust

Power Fuzzy Clustering (PFC) and Power Fuzzy Cluster-wise Regression (PFCR) in R, with Minkowski distances (p = 1 via IRLS, p > 1 via coordinate-wise root finding) and an adaptive, regularised Mahalanobis distance.

Installation

# from a local clone
devtools::install("path/to/pfclust")

Usage

library(pfclust)

# Clustering
res <- PFC(iris[, 1:4], K = 3)
table(res$l, iris[, 5])

# Cluster-wise regression
library(flexCWM)
data("students")
res <- PFCR(students[, 4], students[, 2:3], K = 2, distance = "Mahalanobis")
table(res$l, students[, 1])

See inst/examples/demo.R for a full walkthrough.

Regenerating documentation

The NAMESPACE and man/*.Rd files in this repo were written to match what roxygen2 produces. To regenerate them from the roxygen comments:

devtools::document()