AI Paper Atlas

Research map / Learning theory and algorithms / Clustering

Low-Rank Doubly Stochastic Matrix Decomposition for Cluster Analysis

Zhirong Yang, Jukka Corander, Erkki Oja

JMLR 2016 · Learning theory and algorithms › Clustering · approach: Subspace clustering · 20 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 DCD instead of k-meansclustering · MNISTpurity: 0.48 -> 0.97Improved
Use DCD instead of best baseline on large dataclustering · 19 data sets with N > 4500best-method count: 17 and 11 cases in purity and NMIImproved
Use DCD instead of 1-Specclustering · MNISTpurity: 0.88 -> 0.97Improved
Use DCD instead of LSCclustering · 20NGpurity: 0.50 -> 0.62Improved
Use DCD instead of Ncutclustering · TDT2purity: 0.89 -> 0.86Hurt
Use DCD instead of k-meansclustering · CITESEERpurity: 0.68 -> 0.50Hurt
Use DCD instead of LSDclustering · NORBpurity: 0.42 -> 0.35Hurt
Select cluster number by NOSAC residual instead of CH-kmcluster number selection · MNISTselected number of clusters (truth 10): 2 -> 10Improved
Select cluster number by NOSAC residual instead of gap-kmcluster number selection · COIL100selected number of clusters (truth 100): 140 -> 79Improved
Use 5NN instead of 10NN input similarity (ablation)DCD · COIL20selected number of clusters (truth 20): 17 -> 21Improved
Use 5NN instead of 10NN input similarity (ablation)DCD · COIL100selected number of clusters (truth 100): 79 -> 99Improved

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