English

Complexity and Approximation of the Fuzzy K-Means Problem

Machine Learning 2015-12-21 v1 Data Structures and Algorithms

Abstract

The fuzzy KK-means problem is a generalization of the classical KK-means problem to soft clusterings, i.e. clusterings where each points belongs to each cluster to some degree. Although popular in practice, prior to this work the fuzzy KK-means problem has not been studied from a complexity theoretic or algorithmic perspective. We show that optimal solutions for fuzzy KK-means cannot, in general, be expressed by radicals over the input points. Surprisingly, this already holds for very simple inputs in one-dimensional space. Hence, one cannot expect to compute optimal solutions exactly. We give the first (1+ϵ)(1+\epsilon)-approximation algorithms for the fuzzy KK-means problem. First, we present a deterministic approximation algorithm whose runtime is polynomial in NN and linear in the dimension DD of the input set, given that KK is constant, i.e. a polynomial time approximation algorithm given a fixed KK. We achieve this result by showing that for each soft clustering there exists a hard clustering with comparable properties. Second, by using techniques known from coreset constructions for the KK-means problem, we develop a deterministic approximation algorithm that runs in time almost linear in NN but exponential in the dimension DD. We complement these results with a randomized algorithm which imposes some natural restrictions on the input set and whose runtime is comparable to some of the most efficient approximation algorithms for KK-means, i.e. linear in the number of points and the dimension, but exponential in the number of clusters.

Keywords

Cite

@article{arxiv.1512.05947,
  title  = {Complexity and Approximation of the Fuzzy K-Means Problem},
  author = {Johannes Blömer and Sascha Brauer and Kathrin Bujna},
  journal= {arXiv preprint arXiv:1512.05947},
  year   = {2015}
}
R2 v1 2026-06-22T12:13:18.129Z