Research map / Learning theory and algorithms / Clustering
Covariance-based Clustering in Multivariate and Functional Data Analysis
Francesca Ieva, Anna Maria Paganoni, Nicholas Tarabelloni
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 covariance-based Max-Swap clustering instead of K-means | clustering · Hourglass (bivariate) | misclassification rate: 0.47 ± 0.01 -> 3 of 200 misclassified per cluster | Improved |
| use covariance-based Max-Swap clustering instead of K-means | clustering · Bull's eye (bivariate) | misclassification rate: 0.27 ± 0.01 -> 0 misclassified | Improved |
| use Max-Swap instead of hierarchical clustering with Ward linkage | clustering · Hourglass (bivariate) | misclassification rate: 0.46 ± 0.04 -> 3 of 200 misclassified per cluster | Improved |
| add linear shrinkage covariance estimator instead of sample covariance (ablation) | Max-Swap · Synthetic functional data N=40 | clustering error (%): 5% -> 0% | Improved |
| add linear shrinkage covariance estimator instead of sample covariance (ablation) | Max-Swap · Synthetic functional data N=70 | clustering error (%): 8.5% -> 3% | Improved |
| add linear shrinkage covariance estimator instead of sample covariance (ablation) | Max-Swap · Synthetic functional data N=80 | clustering error (%): 2.5% -> 2.5% | No clear effect |
| add linear shrinkage covariance estimator instead of sample covariance (ablation) | Max-Swap · Synthetic functional data N=60 | clustering error (%): 0% -> 0% | No clear effect |
| increase sample size N from 40 to 100 (ablation) | Max-Swap · Synthetic functional data | clustering error (%): 5% -> 0% | Improved |
| use shrinkage estimator on hemoglobin data instead of sample covariance (ablation) | Max-Swap · Deoxygenated hemoglobin data (N=30) | partition | No clear effect |
Closest papers by meaning
- Shared Subspace Models for Multi-Group Covariance Estimation (JMLR 2019 · similarity 0.89)
- Achieving Optimal Clustering in Gaussian Mixture Models with Anisotropic Covariance Structures (NeurIPS 2024 · similarity 0.87)
- A Non–Convex Optimization Approach to Correlation Clustering (AAAI 2019 · similarity 0.86)
- A Unified Framework for Variable Selection in Model-Based Clustering with Missing Not at Random (NeurIPS 2025 · similarity 0.86)
- A Well-Conditioned and Sparse Estimation of Covariance and Inverse Covariance Matrices Using a Joint Penalty (JMLR 2016 · similarity 0.86)
- Global-Local Dirichlet Processes for Clustering Grouped Data in the Presence of Group-Specific Idiosyncratic Variables (ICML 2025 · similarity 0.86)
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.
