Research map / Label-efficient and robust learning / Recognition and long-tail
A Human-Like Semantic Cognition Network for Aspect-Level Sentiment Classification
Zeyang Lei, Yujiu Yang, Min Yang, Wei Zhao, Jun Guo, Yi 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 |
|---|---|---|---|
| Use HSCN instead of RAM (average results) | aspect-level sentiment classifier · Laptop | accuracy: 72.1 -> 74.9 | Improved |
| Use HSCN instead of RAM (average results) | aspect-level sentiment classifier · Laptop | Macro-F1: 68.4 -> 70.9 | Improved |
| Use HSCN instead of RAM (average results) | aspect-level sentiment classifier · Restaurant | Macro-F1: 68.5 -> 70.0 | Improved |
| Use HSCN instead of SA-LSTM-P | aspect-level sentiment classifier · Restaurant | accuracy: 81.6 -> 77.8 | Hurt |
| Use HSCN instead of RAM | aspect-level sentiment classifier · Tweets | Macro-F1: 67.3 -> 66.1 | Hurt |
| Add word-level interactive perception module (pre-reading) (ablation) | HSCN · SemEval-14 / Tweets | accuracy | Improved |
| Add skip-reading module (ablation) | HSCN · SemEval-14 / Tweets | accuracy | Improved |
| Add semantic composition module (ablation) | HSCN · SemEval-14 / Tweets | accuracy | Improved |
| Add semantic feedback module (rereading) (ablation) | HSCN · SemEval-14 / Tweets | accuracy | Improved |
| Increase readings from 1 to 2 (ablation) | HSCN · Laptop | accuracy: 75.6 -> 76.1 | Improved |
| Increase readings from 2 to 3 (ablation) | HSCN · Laptop | accuracy: 76.1 -> 76.1 | No clear effect |
| Increase readings from 3 to 4 (ablation) | HSCN · Laptop | accuracy: 76.1 -> 75.5 | Hurt |
Closest papers by meaning
- Learning Latent Opinions for Aspect-level Sentiment Classification (AAAI 2018 · similarity 0.86)
- Replicate, Walk, and Stop on Syntax: An Effective Neural Network Model for Aspect-Level Sentiment Classification (AAAI 2020 · similarity 0.86)
- Exploiting Coarse-to-Fine Task Transfer for Aspect-Level Sentiment Classification (AAAI 2019 · similarity 0.85)
- Learning to Attend via Word-Aspect Associative Fusion for Aspect-Based Sentiment Analysis (AAAI 2018 · similarity 0.85)
- You Only Read Once: Constituency-Oriented Relational Graph Convolutional Network for Multi-Aspect Multi-Sentiment Classification (AAAI 2024 · similarity 0.84)
- Modeling Adaptive Inter-Task Feature Interactions via Sentiment-Aware Contrastive Learning for Joint Aspect-Sentiment Prediction (AAAI 2024 · similarity 0.84)
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.
