Research map / Label-efficient and robust learning / Domain adaptation and generalization
Recurrent Neural Networks with Auxiliary Labels for Cross-Domain Opinion Target Extraction
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 |
|---|---|---|---|
| Hier-Joint with rule-based auxiliary labels instead of direct source training | Direct-2 LSTM · Cross-domain OTE average (12 pairs) | F1: 0.281 -> 0.342 | Helped |
| joint instead of separate training (ablation) | Hierarchical network · Cross-domain OTE average (12 pairs) | F1: 0.297 -> 0.342 | Helped |
| joint instead of separate training (ablation) | Concatenated network · Cross-domain OTE average (12 pairs) | F1: 0.265 -> 0.280 | Helped |
| hierarchical instead of concatenated network (ablation) | joint training model · Cross-domain OTE average (12 pairs) | F1: 0.280 -> 0.342 | Helped |
| naively mixing auxiliary-labeled target data with source data | Direct LSTM · Cross-domain OTE average (12 pairs) | F1: 0.265 -> 0.252 | Hurt |
| LSTM instead of RNN/BiRNN/BiLSTM (ablation) | Hier-Joint · Device target | F1 | Helped |
| hidden dim beyond 100 (ablation) | Hier-Joint · Device target | F1 | Hurt |
| prune rules by source-domain precision (ablation) | Hier-Joint · Device target | F1 | Mixed |
Closest papers by meaning
- A Unified Model for Opinion Target Extraction and Target Sentiment Prediction (AAAI 2019 · similarity 0.88)
- Transferable Interactive Memory Network for Domain Adaptation in Fine-Grained Opinion Extraction (AAAI 2019 · similarity 0.85)
- Latent Opinions Transfer Network for Target-Oriented Opinion Words Extraction (AAAI 2020 · similarity 0.85)
- An Adaptive Hybrid Framework for Cross-domain Aspect-based Sentiment Analysis (AAAI 2021 · similarity 0.83)
- Automatic Rule Extraction from Long Short Term Memory Networks (ICLR 2017 · similarity 0.83)
- Gated Neural Networks for Targeted Sentiment Analysis (AAAI 2016 · similarity 0.83)
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.