Research map / Probabilistic and scientific modeling / Molecules and proteins
MCbiF: Measuring Topological Autocorrelation in Multiscale Clusterings via 2-Parameter Persistent Homology
Juni Schindler, Mauricio Barahona
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 MCbiF Hilbert-function features instead of the GCN on Sankey graphs | crossing-number regression · Sankey crossing-number regression (N=5) | test R2: 0.416 -> 0.539 (LR on HF0&HF1) | Improved |
| use combined HF0 & HF1 instead of best baseline CE | crossing-number regression · Sankey crossing-number regression (N=10) | test R2: 0.458 -> 0.516 | Improved |
| add HF0 features on top of HF1 (ablation) | MCbiF feature maps (LR) · Sankey crossing-number regression (N=5) | test R2: 0.486 -> 0.539 | Improved |
| use HF1 instead of HF0 alone (ablation) | MCbiF feature maps (LR) · Sankey crossing-number regression (N=5) | test R2: 0.147 -> 0.486 | Improved |
| use HF1 features instead of CE, ARI, MOD or raw label encoding | logistic regression · Order-preservation classification (N=500, M=30) | test accuracy: 0.53 (raw) -> 0.97 | Improved |
| use HF0 instead of HF1 for order-preservation detection (ablation) | logistic regression · Order-preservation classification (N=500, M=30) | test accuracy: 0.97 -> 0.56 | Hurt |
| use nerve-based MCbiF instead of the original MCbiF (ablation) | mice contact data · Mice contact data | simplices / computation time: 116,700 -> 260 simplices; 50-fold time reduction | Improved |
| use raw label encoding with a representation learner (CNN) at N=5 vs N=10 | raw label encoding baseline · Sankey crossing-number regression | test R2: 0.267 -> 0.072 (CNN, N=5 -> N=10) | Hurt |
Closest papers by meaning
- Stable and Consistent Density-Based Clustering via Multiparameter Persistence (JMLR 2024 · similarity 0.90)
- ToMAToMP: Robust and Multi-Parameter Topological Clustering (NeurIPS 2026 · similarity 0.89)
- Learning Persistent Community Structures in Dynamic Networks via Topological Data Analysis (AAAI 2024 · similarity 0.87)
- Graphcode: Learning from multiparameter persistent homology using graph neural networks (NeurIPS 2024 · similarity 0.86)
- Persistent Homology for High-dimensional Data Based on Spectral Methods (NeurIPS 2024 · similarity 0.85)
- Classification Fields: Arbitrarily Fine Recursive Hierarchical Clustering From Few Examples (NeurIPS 2026 · similarity 0.85)
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.
