Research map / Learning theory and algorithms / Clustering
Feature Sampling Based Unsupervised Semantic Clustering for Real Web Multi-View Content
Xiaolong Gong, Linpeng Huang, Fuwei Wang
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 |
|---|---|---|---|
| remove inter-semantic diversity (HSIC) term, keep only intra-semantic similarity (FSUSC-intra) (ablation) | FSUSC · Doucom | ACC: 43.2 -> 58.6 | Improved |
| remove intra-semantic similarity term, keep only inter-semantic diversity (FSUSC-inter) (ablation) | FSUSC · Doucom | ACC: 52.5 -> 58.6 | Improved |
| remove intra-semantic similarity term (FSUSC-inter) (ablation) | FSUSC · 3-Sources | ACC: 73.2 -> 74.1 | Improved |
| remove inter-semantic diversity term (FSUSC-intra) (ablation) | FSUSC · Yelp | NMI: 63.2 -> 67.2 | Improved |
| replace unsupervised semantic similarity with supervised label similarity (ablation) | FSUSC · Doucom | ACC: 45.6 -> 58.6 | Improved |
| replace unsupervised semantic similarity with supervised label similarity (ablation) | FSUSC · 3-Sources | NMI: 67.9 -> 72.6 | Improved |
| use Gaussian kernel (GK) similarity on whole feature space instead of semantic (SC) topic similarity (ablation) | FSUSC · Doucom | ACC: 50.4 -> 58.6 | Improved |
| use Gaussian kernel similarity instead of semantic topic similarity (ablation) | FSUSC · Last.fm | NMI: 56.4 -> 61.7 | Improved |
| use Gaussian kernel similarity instead of semantic topic similarity (ablation) | FSUSC · 3-Sources | ACC: 72.8 -> 74.1 | Improved |
| random feature sampling instead of fixed top sampling (ablation) | FSUSC · Yelp and 3-Sources | ACC and NMI | Improved |
| feature sampling with reduction factor eta=8 versus full-feature updates | FSUSC · Doucom | convergence time (s): 30,981 -> 10,732 | Improved |
| feature sampling versus full features (compare with MultiNMF) | FSUSC · Last.fm | convergence time (s): 7,768 -> 2,771 | Improved |
Closest papers by meaning
- A Non-parametric Graph Clustering Framework for Multi-View Data (AAAI 2024 · similarity 0.87)
- Sample-Level Cross-View Similarity Learning for Incomplete Multi-View Clustering (AAAI 2024 · similarity 0.87)
- Medusa: A Multi-Scale High-order Contrastive Dual-Diffusion Approach for Multi-View Clustering (CVPR 2025 · similarity 0.86)
- Fast Incomplete Multi-view Clustering with Adaptive Similarity Completion and Reconstruction (AAAI 2025 · similarity 0.86)
- Incomplete and Unpaired Multi-View Graph Clustering with Cross-View Feature Fusion (AAAI 2025 · similarity 0.86)
- Collaborative Similarity Fusion and Consistency Recovery for Incomplete Multi-view Clustering (AAAI 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.
