Research map / Label-efficient and robust learning / Semi-supervised learning
Semi-Supervised Multinomial Naive Bayes for Text Classification by Leveraging Word-Level Statistical Constraint
Li Zhao, Minlie Huang, Ziyu Yao, Rongwei Su, Yingying Jiang, Xiaoyan Zhu
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 word-level statistical constraint (MNB-WSC) (ablation) | MNB-EM · Ohscal, 512 labeled | Macro-F1: 0.6404 -> 0.6736 | Improved |
| add word-level statistical constraint (MNB-WSC) (ablation) | MNB-EM · 20News, 64 labeled | Macro-F1: 0.4242 -> 0.4429 | Improved |
| add word-level statistical constraint (MNB-WSC) (ablation) | MNB-EM · Reuters, 512 labeled | Macro-F1: 0.7714 -> 0.7723 | No clear effect |
| add word-level statistical constraint (MNB-WSC) (ablation) | MNB-EM · WebKB, 512 labeled | Macro-F1: 0.7641 -> 0.7839 | Improved |
| add word-level statistical constraint (MNB-WSC) (ablation) | MNB-EM · Kitchen sentiment, 512 labeled | Macro-F1: 0.6158 -> 0.7182 | Improved |
| add word-level statistical constraint (MNB-WSC) (ablation) | MNB-EM · Dvd sentiment, 64 labeled | Macro-F1: 0.4827 -> 0.5546 | Improved |
| vary confidence level of Wilson interval from 70% to 95% (ablation) | MNB-EM · Ohscal, 512 labeled | Macro-F1 | No clear effect |
| increase labeled set size from 64 to 512 for MNB-WSC (ablation) | MNB-EM · Ohscal | Macro-F1: 0.5417 -> 0.6736 | Improved |
| use MNB-WSC instead of plain MNB-EM in the KL divergence of word class distribution (ablation) | MNB-EM · Kitchen, 64 labeled | KL-divergence from true word class distribution | Improved |
Closest papers by meaning
- SimPro: A Simple Probabilistic Framework Towards Realistic Long-Tailed Semi-Supervised Learning (ICML 2024 · similarity 0.85)
- Variational Bayesian Semi-Supervised Keyword Extraction (TPAMI 2025 · similarity 0.84)
- Wasserstein Gradient Boosting: A Framework for Distribution-Valued Supervised Learning (NeurIPS 2024 · similarity 0.80)
- Improved Graph Contrastive Learning for Short Text Classification (AAAI 2024 · similarity 0.80)
- Boosting Short Text Classification with Multi-Source Information Exploration and Dual-Level Contrastive Learning (AAAI 2025 · similarity 0.80)
- Compositional Generalization for Multi-Label Text Classification: A Data-Augmentation Approach (AAAI 2024 · similarity 0.80)
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.
