TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Papers/Utilizing BERT for Aspect-Based Sentiment Analysis via Con...

Utilizing BERT for Aspect-Based Sentiment Analysis via Constructing Auxiliary Sentence

Chi Sun, Luyao Huang, Xipeng Qiu

2019-03-22NAACL 2019 6Question AnsweringSentiment AnalysisNatural Language InferenceAspect-Based Sentiment AnalysisAspect-Based Sentiment Analysis (ABSA)Sentence-Pair ClassificationGeneral Classification
PaperPDFCodeCodeCodeCodeCodeCodeCodeCode(official)

Abstract

Aspect-based sentiment analysis (ABSA), which aims to identify fine-grained opinion polarity towards a specific aspect, is a challenging subtask of sentiment analysis (SA). In this paper, we construct an auxiliary sentence from the aspect and convert ABSA to a sentence-pair classification task, such as question answering (QA) and natural language inference (NLI). We fine-tune the pre-trained model from BERT and achieve new state-of-the-art results on SentiHood and SemEval-2014 Task 4 datasets.

Results

TaskDatasetMetricValueModel
Sentiment AnalysisSemEval 2014 Task 4 Subtask 4Accuracy (3-way)89.9BERT-pair-QA-B
Sentiment AnalysisSemEval 2014 Task 4 Subtask 4Accuracy (4-way)85.9BERT-pair-QA-B
Sentiment AnalysisSemEval 2014 Task 4 Subtask 4Binary Accuracy95.6BERT-pair-QA-B
Sentiment AnalysisSentihoodAspect87.9BERT-pair-QA-B
Sentiment AnalysisSentihoodSentiment93.3BERT-pair-QA-B
Sentiment AnalysisSentihoodAspect86.4BERT-pair-QA-M
Sentiment AnalysisSentihoodSentiment93.6BERT-pair-QA-M
Aspect Category DetectionSemEval 2014 Task 4 Subtask 3F1 score92.18BERT-pair-NLI-B
Aspect Category DetectionSemEval 2014 Task 4 Subtask 3Precision93.57BERT-pair-NLI-B
Aspect Category DetectionSemEval 2014 Task 4 Subtask 3Recall90.83BERT-pair-NLI-B
Aspect-Based Sentiment Analysis (ABSA)SemEval 2014 Task 4 Subtask 4Accuracy (3-way)89.9BERT-pair-QA-B
Aspect-Based Sentiment Analysis (ABSA)SemEval 2014 Task 4 Subtask 4Accuracy (4-way)85.9BERT-pair-QA-B
Aspect-Based Sentiment Analysis (ABSA)SemEval 2014 Task 4 Subtask 4Binary Accuracy95.6BERT-pair-QA-B
Aspect-Based Sentiment Analysis (ABSA)SentihoodAspect87.9BERT-pair-QA-B
Aspect-Based Sentiment Analysis (ABSA)SentihoodSentiment93.3BERT-pair-QA-B
Aspect-Based Sentiment Analysis (ABSA)SentihoodAspect86.4BERT-pair-QA-M
Aspect-Based Sentiment Analysis (ABSA)SentihoodSentiment93.6BERT-pair-QA-M

Related Papers

From Roots to Rewards: Dynamic Tree Reasoning with RL2025-07-17Enter the Mind Palace: Reasoning and Planning for Long-term Active Embodied Question Answering2025-07-17Vision-and-Language Training Helps Deploy Taxonomic Knowledge but Does Not Fundamentally Alter It2025-07-17City-VLM: Towards Multidomain Perception Scene Understanding via Multimodal Incomplete Learning2025-07-17AdaptiSent: Context-Aware Adaptive Attention for Multimodal Aspect-Based Sentiment Analysis2025-07-17Describe Anything Model for Visual Question Answering on Text-rich Images2025-07-16Is This Just Fantasy? Language Model Representations Reflect Human Judgments of Event Plausibility2025-07-16AI Wizards at CheckThat! 2025: Enhancing Transformer-Based Embeddings with Sentiment for Subjectivity Detection in News Articles2025-07-15