# Aspect-Based Sentiment Analysis on Rest 14 (Accuracy, Macro-F1) Source: [Aspect-Based Sentiment Analysis on Rest 14 (Accuracy, Macro-F1)]() Collection index: [llms.txt]() Data updated: 2026-05-21T12:21:52.263Z Evaluation of Aspect-Based Sentiment Analysis performance on the Rest 14 dataset, reporting Accuracy and Macro-F1 scores. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Aspect-Based Sentiment Analysis | | Dataset (including any stated split) | Rest 14 | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 31 of 31 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Accuracy](): Higher is better; unit: PERCENT. - [Macro-F1](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | DABS | 89.76 | Not recorded | 6a0e91b2f984739ed6738c35 | | DeBERTa+RCL | 89.38 | Not recorded | 6a0e91b2f984739ed6738c35 | | PConvRoBERTa | 89.29 | Not recorded | 6a0e91b2f984739ed6738c35 | | Flan-T5-base+Syn-Chain | 88.39 | Not recorded | 6a0e91b2f984739ed6738c35 | | ITGCN | 88.21 | Not recorded | 6a0e91b2f984739ed6738c35 | | LWEDA-ACT (2025) | 88.11 | Not recorded | 68fa7b4a1e209131cea87957 | | RDGCN (2024) | 87.49 | Not recorded | 68fa7b4a1e209131cea87957 | | CSADGCN (2024) | 87.4 | Not recorded | 68fa7b4a1e209131cea87957 | | DSSK-GAN-BERT | 87.32 | Not recorded | 6a0e91b2f984739ed6738c35 | | SSEGCN (2022) | 87.31 | Not recorded | 68fa7b4a1e209131cea87957 | | DC-GCN | 87.29 | Not recorded | 6a0e91b2f984739ed6738c35 | | IDGNN (2024) | 87.25 | Not recorded | 68fa7b4a1e209131cea87957 | | OTESGN | 87.21 | Not recorded | 68fa7b4a1e209131cea87957 | | KGAN-BERT | 87.15 | Not recorded | 6a0e91b2f984739ed6738c35 | | CADA (2025) | 87.14 | Not recorded | 68fa7b4a1e209131cea87957 | | YORO (2024) | 87.14 | Not recorded | 68fa7b4a1e209131cea87957 | | DualGCN (2021) | 87.13 | Not recorded | 68fa7b4a1e209131cea87957 | | KDGN (2023) | 87.01 | Not recorded | 68fa7b4a1e209131cea87957 | | APSCL (2023) | 86.86 | Not recorded | 68fa7b4a1e209131cea87957 | | HPEP-GCN (2025) | 86.86 | Not recorded | 68fa7b4a1e209131cea87957 | | SVCCL (2025) | 86.34 | Not recorded | 68fa7b4a1e209131cea87957 | | DGEDT (2020) | 86.3 | Not recorded | 68fa7b4a1e209131cea87957 | | BERT (2019) | 85.97 | Not recorded | 68fa7b4a1e209131cea87957 | | CABiLSTM-BERT | 83.75 | Not recorded | 6a0e91b2f984739ed6738c35 | | ASHGAT | 83.45 | Not recorded | 6a0e91b2f984739ed6738c35 | | Llama-3-8B-Inst | 82.87 | Not recorded | 6a0e91b2f984739ed6738c35 | | Qwen3-8B | 82.75 | Not recorded | 6a0e91b2f984739ed6738c35 | | ASGCN (2019) | 80.77 | Not recorded | 68fa7b4a1e209131cea87957 | | TNet (2018) | 80.69 | Not recorded | 68fa7b4a1e209131cea87957 | | RAM (2017) | 80.23 | Not recorded | 68fa7b4a1e209131cea87957 | | GPT-3.5 Turbo | 78.85 | Not recorded | 6a0e91b2f984739ed6738c35 | ### DABS Accuracy: 89.76. Result ID: 6a0ef8cbfbc373592c07794c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Seeds":"3","Evaluation protocol":"Fine-tuning"}. This record does not assert independent reproduction. ### DeBERTa+RCL Accuracy: 89.38. Result ID: 6a0ef8cbfbc373592c077947. [Reporting paper]() Recorded evaluation configuration: {"Evaluation protocol":"Fine-tuning"}. This record does not assert independent reproduction. ### PConvRoBERTa Accuracy: 89.29. Result ID: 6a0ef8cbfbc373592c077945. [Reporting paper]() Recorded evaluation configuration: {"Evaluation protocol":"Fine-tuning"}. This record does not assert independent reproduction. ### Flan-T5-base+Syn-Chain Accuracy: 88.39. Result ID: 6a0ef8cbfbc373592c077948. [Reporting paper]() Recorded evaluation configuration: {"Evaluation protocol":"Fine-tuning"}. This record does not assert independent reproduction. ### ITGCN Accuracy: 88.21. Result ID: 6a0ef8cbfbc373592c077946. [Reporting paper]() Recorded evaluation configuration: {"Evaluation protocol":"Fine-tuning"}. This record does not assert independent reproduction. ### LWEDA-ACT (2025) Accuracy: 88.11. Result ID: 69b009e66dbb7075b959bb49. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RDGCN (2024) Accuracy: 87.49. Result ID: 69b009e66dbb7075b959bb40. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### CSADGCN (2024) Accuracy: 87.4. Result ID: 69b009e66dbb7075b959bb41. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DSSK-GAN-BERT Accuracy: 87.32. Result ID: 6a0ef8cbfbc373592c077942. [Reporting paper]() Recorded evaluation configuration: {"Evaluation protocol":"Fine-tuning"}. This record does not assert independent reproduction. ### SSEGCN (2022) Accuracy: 87.31. Result ID: 69b009e66dbb7075b959bb3f. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DC-GCN Accuracy: 87.29. Result ID: 6a0ef8cbfbc373592c077943. [Reporting paper]() Recorded evaluation configuration: {"Evaluation protocol":"Fine-tuning"}. This record does not assert independent reproduction. ### IDGNN (2024) Accuracy: 87.25. Result ID: 69b009e66dbb7075b959bb47. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### OTESGN Accuracy: 87.21. Result ID: 69b009e66dbb7075b959bb4a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### KGAN-BERT Accuracy: 87.15. Result ID: 6a0ef8cbfbc373592c077940. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation protocol":"Fine-tuning"}. This record does not assert independent reproduction. ### CADA (2025) Accuracy: 87.14. Result ID: 69b009e66dbb7075b959bb44. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### YORO (2024) Accuracy: 87.14. Result ID: 69b009e66dbb7075b959bb46. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DualGCN (2021) Accuracy: 87.13. Result ID: 69b009e66dbb7075b959bb3e. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### KDGN (2023) Accuracy: 87.01. Result ID: 69b009e66dbb7075b959bb45. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### APSCL (2023) Accuracy: 86.86. Result ID: 69b009e66dbb7075b959bb42. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### HPEP-GCN (2025) Accuracy: 86.86. Result ID: 69b009e66dbb7075b959bb48. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### SVCCL (2025) Accuracy: 86.34. Result ID: 69b009e66dbb7075b959bb43. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DGEDT (2020) Accuracy: 86.3. Result ID: 69b009e66dbb7075b959bb3d. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### BERT (2019) Accuracy: 85.97. Result ID: 69b009e66dbb7075b959bb3b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### CABiLSTM-BERT Accuracy: 83.75. Result ID: 6a0ef8cbfbc373592c077944. [Reporting paper]() Recorded evaluation configuration: {"Evaluation protocol":"Fine-tuning"}. This record does not assert independent reproduction. ### ASHGAT Accuracy: 83.45. Result ID: 6a0ef8cbfbc373592c077941. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation protocol":"Fine-tuning"}. This record does not assert independent reproduction. ### Llama-3-8B-Inst Accuracy: 82.87. Result ID: 6a0ef8cbfbc373592c077949. [Reporting paper]() Recorded evaluation configuration: {"Number of shots":"5-shot"}. This record does not assert independent reproduction. ### Qwen3-8B Accuracy: 82.75. Result ID: 6a0ef8cbfbc373592c07794a. [Reporting paper]() Recorded evaluation configuration: {"Number of shots":"5-shot"}. This record does not assert independent reproduction. ### ASGCN (2019) Accuracy: 80.77. Result ID: 69b009e66dbb7075b959bb3c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### TNet (2018) Accuracy: 80.69. Result ID: 69b009e66dbb7075b959bb3a. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RAM (2017) Accuracy: 80.23. Result ID: 69b009e66dbb7075b959bb39. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GPT-3.5 Turbo Accuracy: 78.85. Result ID: 6a0ef8cbfbc373592c07794b. [Reporting paper]() Recorded evaluation configuration: {"Number of shots":"5-shot"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.