Research map / Label-efficient and robust learning / Recognition and long-tail
Improving Review Representations With User Attention and Product Attention for Sentiment Classification
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| combine separate user-attention and product-attention representations (ablation) | HUAPA · IMDB | accuracy: 0.493 -> 0.550 | Helped |
| combine separate user-attention and product-attention representations (ablation) | HUAPA · Yelp 2013 | accuracy: 0.641 -> 0.683 | Helped |
| combine separate user-attention and product-attention representations (ablation) | HUAPA · Yelp 2014 | accuracy: 0.646 -> 0.686 | Helped |
| add user-view auxiliary loss (ablation) | HUAPA · Yelp 2013 | accuracy: 0.669 -> 0.672 | Helped |
| add product-view auxiliary loss (ablation) | HUAPA · Yelp 2013 | accuracy: 0.669 -> 0.675 | Helped |
| add both user-view and product-view auxiliary losses (ablation) | HUAPA · IMDB | accuracy: 0.538 -> 0.550 | Helped |
| add both user-view and product-view auxiliary losses (ablation) | HUAPA · Yelp 2013 | accuracy: 0.669 -> 0.683 | Helped |
| add both user-view and product-view auxiliary losses (ablation) | HUAPA · Yelp 2014 | accuracy: 0.675 -> 0.686 | Helped |
| use user attention only, without product attention (ablation) | HUAPA · IMDB | accuracy: 0.550 -> 0.521 | Hurt |
| use product attention only, without user attention (ablation) | HUAPA · IMDB | accuracy: 0.550 -> 0.493 | Hurt |
| use user attention only, without product attention (ablation) | HUAPA · Yelp 2013 | accuracy: 0.683 -> 0.649 | Hurt |
| use product attention only, without user attention (ablation) | HUAPA · Yelp 2013 | accuracy: 0.683 -> 0.641 | Hurt |
Closest papers by meaning
- SNNN: Promoting Word Sentiment and Negation in Neural Sentiment Classification (AAAI 2018 · similarity 0.86)
- Asymmetrical Hierarchical Networks with Attentive Interactions for Interpretable Review-Based Recommendation (AAAI 2020 · similarity 0.85)
- MNCN: A Multilingual Ngram-Based Convolutional Network for Aspect Category Detection in Online Reviews (AAAI 2019 · similarity 0.84)
- Probing Product Description Generation via Posterior Distillation (AAAI 2021 · similarity 0.83)
- UGSD: User Generated Sentiment Dictionaries from Online Customer Reviews (AAAI 2019 · similarity 0.83)
- Review-Enhanced Hierarchical Contrastive Learning for Recommendation (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.