Research map / Label-efficient and robust learning / Recognition and long-tail
Unsupervised Personalized Feature Selection
Jundong Li, Liang Wu, Harsh Dani, Huan Liu
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 |
|---|---|---|---|
| add instance-specific localized weights, exclusive group lasso and network lasso to NDFS (global weights only) | UPFS · BlogCatalog | clustering ACC (%): 26.51 -> 28.30 | Improved |
| add personalized feature weights over NDFS | UPFS · DBLP | clustering ACC (%): 35.60 -> 39.72 | Improved |
| add personalized feature weights over NDFS | UPFS · CNNStory | clustering NMI (%): 51.63 -> 59.71 | Improved |
| add personalized feature weights over NDFS | UPFS · Yale | clustering ACC (%): 42.48 -> 39.92 | Hurt |
| add personalized feature weights over NDFS | UPFS · warpPIE10P | clustering ACC (%): 37.60 -> 28.74 | Hurt |
| add personalized feature weights over NDFS | UPFS · Carcinoma | clustering ACC (%): 73.11 -> 76.98 | Improved |
| add personalized feature weights over NDFS | UPFS · TOX-171 | clustering ACC (%): 47.89 -> 49.98 | Improved |
| select features with UPFS versus using all features | UPFS · CNNStory | clustering ACC (%): 50.70 -> 54.82 | Improved |
| select features with UPFS versus using all features | UPFS · warpPIE10P | clustering ACC (%): 26.67 -> 28.74 | Improved |
| vary alpha, beta, gamma regularization parameters in 0.1 to 10 (ablation) | UPFS · BlogCatalog | clustering ACC (%) | No clear effect |
Closest papers by meaning
- Unsupervised Feature Selection Through Group Discovery (AAAI 2026 · similarity 0.85)
- Consensus Guided Unsupervised Feature Selection (AAAI 2016 · similarity 0.84)
- Most Influential Subset Selection: Challenges, Promises, and Beyond (NeurIPS 2024 · similarity 0.82)
- SAND: One-Shot Feature Selection with Additive Noise Distortion (ICML 2025 · similarity 0.81)
- Local Feature Selection without Label or Feature Leakage for Interpretable Machine Learning Predictions (ICML 2024 · similarity 0.81)
- Selective Deep Autoencoder for Unsupervised Feature Selection (AAAI 2024 · similarity 0.81)
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.
