AI Paper Atlas

Research map / Learning theory and algorithms / Clustering

A Flexible EM-Like Clustering Algorithm for Noisy Data

Violeta Roizman, Matthieu Jonckheere, Frédéric Pascal

TPAMI 2023 · Learning theory and algorithms › Clustering · approach: Clustering approximation algorithms · 29 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use F-EM instead of GMM-EMclustering · 20newsgroupAMI: 0.3526 -> 0.4873Improved
use F-EM instead of GMM-EMclustering · small NORBAMI: 0.0476 -> 0.5321Improved
use F-EM instead of t-EMclustering · small NORBAMI: 0.4370 -> 0.5321Improved
use F-EM instead of t-EMclustering · MNIST 7-1AMI: 0.8947 -> 0.8811Hurt
use F-EM instead of spectral clusteringclustering · MNIST 3-8-6AMI: 0.8272 -> 0.7918Hurt
use F-EM instead of TCLUSTclustering · MNIST 3-8-6 + noiseAMI: 0.6047 -> 0.4664Hurt
use F-EM instead of spectral clusteringclustering · MNIST 3-8-6 + noiseAMI: 0.3511 -> 0.4664Improved
use F-EM instead of GMM-EMclustering · Synthetic setup 5 (mixed distributions per cluster)ARI: 0.0187 -> 0.6946Improved
use F-EM instead of GMM-EMclustering · Synthetic setup 3 (K, t, Gaussian clusters)ARI: 0.7056 -> 0.9722Improved
use F-EM instead of t-EMclustering · Synthetic setup 1 (t, dof=3)ARI: 0.7895 -> 0.7513Hurt
use F-EM instead of t-EMclustering · Synthetic setup 2 (t, dof=10)ARI: 0.9233 -> 0.9208No clear effect
use F-EM instead of TCLUSTclustering · Synthetic setup 4 (uniform noise)ARI: 0.9690 -> 0.8159Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy