Research map / Label-efficient and robust learning / Semi-supervised learning
Minimum Density Hyperplanes
Nicos G. Pavlidis, David P. Hofmeyr, Sotiris K. Tasoulis
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use MDP2 low-density hyperplane instead of dePDDP for binary partitioning | hyperplane clustering · 14 benchmark data sets | average improvement of MDP2 (SR, V-m): 0.40 (SR), 0.32 (V-m) | Improved |
| use MDP2 instead of spectral clustering SCn | hyperplane clustering · 14 benchmark data sets | SR and V-m: 0.22 (SR), 0.16 (V-m) | Mixed |
| search for optimised projections (MDP2) instead of always using the first principal component | PDDP-style divisive clustering · 14 benchmark data sets | V-measure: 0.11 (SR), 0.18 (V-m) | Improved |
| use MDP2 instead of 2-means++ | clustering · 14 benchmark data sets | SR and V-m: 0.10 (SR), 0.08 (V-m) | Improved |
| use MDP2 instead of iSVR maximum margin clustering | clustering · 14 benchmark data sets | SR and V-m: 0.13 (SR), 0.18 (V-m) | Improved |
| use MDP2 instead of LapSVM, SSSL, XNV with few labelled examples | semi-supervised linear classification · eight UCI data sets, 5 to 100 labelled examples | classification error on unlabelled data | Improved |
| use MDP2 instead of LapSVM on ionosphere | semi-supervised classification · ionosphere | classification error | Hurt |
| increase the number of labelled examples up to 300 for MDP2 | MDP2 semi-supervised · multiclass data sets | classification error | Mixed |
Closest papers by meaning
- Clustering by Minimum Cut Hyperplanes (TPAMI 2016 · similarity 0.89)
- Hyperplane Clustering via Dual Principal Component Pursuit (ICML 2017 · similarity 0.89)
- Robust Discriminative Clustering with Sparse Regularizers (JMLR 2017 · similarity 0.87)
- Semi-Supervised Sparse Gaussian Classification: Provable Benefits of Unlabeled Data (NeurIPS 2024 · similarity 0.85)
- On Rate-Optimal Partitioning Classification from Observable and from Privatised Data (NeurIPS 2026 · similarity 0.84)
- Classification Fields: Arbitrarily Fine Recursive Hierarchical Clustering From Few Examples (NeurIPS 2026 · similarity 0.84)
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.
