/Fine-grained-Sentiment-Analysis

A simple summary of fine-grained sentiment analysis

Apache License 2.0Apache-2.0

Fine-grained Sentiment Analysis

Tasks

  • ABSA

    • ASC: Aspect Sentiment Classification
      • ATSA: Aspect-Term Sentiment Analysis
      • ACSA: Aspect-Category Sentiment Analysis
    • ATE: Aspect Term Extraction
    • OTE: Opinion Term Extraction
    • Aspect-Sentiment Pair Extraction (ASPE)
    • Aspect and Opinion Term Co-Extraction (AOTE)

  • Aspect Sentiment Triplet Extraction (ASTE)

  • Emotion Analysis

Tasks Input Output
ASC sentence, aspect aspect sentiment
ASPE sentence aspect term, aspect sentiment
ASTE sentence aspect term, aspect sentiment, opinion term
ATE sentence aspect term
OTE sentence opinion term
AOTE sentence aspect term, opinion term
Emotion Analysis sentence joy, anger, fear, etc.

Performance (Accuracy / Macro-F1)

ASC Models

LAPTOP-14 REST-14 TWITTER REST-15 REST-16
(2016) TD_LSTM - - 70.8/69.0 - -
(2016) TC-LSTM - - 71.5/69.5 - -
(2016) ATAE-LSTM 68.7/- 77.2/- - - -
(2016) MemNet 72.37/- 0.8095/- - - -
(2017) IAN 72.1/- 78.6/- - - -
(2017) RAM 74.49/71.35 80.23/70.80 69.36/67.30 - -
(2018) MGAN 75.39/72.47 81.25/71.94 72.54/70.81 - -
(2018) TNet-LF 76.01/71.47 80.79/70.84 74.68/73.36 - -
(2018) TNet-AS 76.54/71.75 80.69/71.27 74.97/73.60 - -
(2018) PBAN 74.12/- 81.16/- - - -
(2019) AGDT 75.86/- 82.95/- - - -
(2019) IACapsNet 76.80/73.29 81.79/73.40 75.01/73.81
(2019) ASGCN-DG 75.55/71.05 80.77/72.02 72.15/70.40 79.89/61.89 88.99/67.48
(2019) CDT 77.19/72.99 82.30/74.02 74.66/73.66 - 85.58/69.93
(2019) HGMN 76.67/72.22 82.33/73.34 73.70/72.89 - -

Paper list

ASC

  • [COLING-2016] Effective LSTMs for Target-Dependent Sentiment Classification [paper]

  • [EMNLP-2016] Attention-based LSTM for Aspect-level Sentiment Classification [paper]

  • [EMNLP-2016] Aspect Level Sentiment Classification with Deep Memory Network [paper]

  • [ACL-2017] Recurrent Attention Network on Memory for Aspect Sentiment Analysis [paper]

  • [IJCAI-2017] Interactive Attention Networks for Aspect-Level Sentiment Classification [paper]

  • [SBP-BRiMS-2018] Aspect Level Sentiment Classification with Attention-over-Attention Neural Networks [paper]

  • [ACL-2018] Aspect Based Sentiment Analysis with Gated Convolutional Networks [paper]
  • [EMNLP-2018] Multi-grained Attention Network for Aspect-Level Sentiment Classification [paper]

  • [COLING-2018] A Position-aware Bidirectional Attention Network for Aspect-level Sentiment Analysis [paper]

  • [NAACL-2018] Modeling Inter-Aspect Dependencies for Aspect-Based Sentiment Analysis [paper]

  • [EMNLP-2018] IARM: Inter-Aspect Relation Modeling with Memory Networks in Aspect-Based Sentiment Analysis [paper]

  • [CoNLL-2018] Hierarchical Attention Based Position-aware Network for Aspect-level Sentiment Analysis

    [paper] [code]

  • [COLING-2018] Effective Attention Modeling for Aspect-Level Sentiment Classification [paper]

  • [ACL-2018] Transformation Networks for Target-Oriented Sentiment Classification [paper]

  • [IJCAI-2019] Learn to Select via Hierarchical Gate Mechanism for Aspect-Based Sentiment Analysis [paper]

  • [EMNLP-2019] A Novel Aspect-Guided Deep Transition Model for Aspect Based Sentiment Analysis [paper] [code]

  • [EMNLP-2019] Recognizing Conflict Opinions in Aspect-level Sentiment Classification with Dual Attention Networks [paper] [code]

  • [EMNLP-2019] Capsule Network with Interactive Attention for Aspect-Level Sentiment Classification [paper]

  • [AAAI-2019] A Human-Like Semantic Cognition Network for Aspect-Level Sentiment Classification [paper]

  • [EMNLP-2019] CAN: Constrained Attention Networks for Multi-Aspect Sentiment Analysis [paper]

GNN-based

  • [EMNLP-2019] Aspect-based Sentiment Classification with Aspect-specific Graph Convolutional Networks [paper] [code]

  • [EMNLP-2019] Aspect-Level Sentiment Analysis Via Convolution over Dependency Tree [paper] [code]

  • [EMNLP-2019] Syntax-Aware Aspect Level Sentiment Classification with Graph Attention Networks [paper] [code]
  • [arXiv-2019] Modeling Sentiment Dependencies with Graph Convolutional Networks for Aspect-levelSentiment Classification [paper]

  • [arXiv-2020] Exploiting Typed Syntactic Dependencies for Targeted Sentiment ClassificationUsing Graph Attention Neural Network [paper]

  • [arXiv-2020] Selective Attention Based Graph Convolutional Networks for Aspect-Level Sentiment Classification [paper]

  • [ACL-2020] Relational Graph Attention Network for Aspect-based Sentiment Analysis [paper] [code]

ASPE

  • [AAAI-2019] A Unified Model for Opinion Target Extraction and Target Sentiment Prediction [paper]

  • [EMNLP-2019] Exploiting BERT for End-to-End Aspect-based Sentiment Analysis [paper]

  • [ACL-2019] DOER: Dual Cross-Shared RNN for Aspect Term-Polarity Co-Extraction [paper]

  • [IJCAI-2019] A Span-based Joint Model for Opinion Target Extraction and Target Sentiment Classification [paper]

  • [ACL-2020] Modelling Context and Syntactical Features for Aspect-based Sentiment Analysis [paper]

ASTE

  • [AAAI-2020] Knowing What, How and Why: A Near Complete Solution for Aspect-based Sentiment Analysis [paper]

ATE

  • [IJCAI-2018] Aspect Term Extraction with History Attention and Selective Transformation [paper]

  • [ACL-2018] Double Embeddings and CNN-based Sequence Labeling for Aspect Extraction [paper] [code]

  • [ACL-2019] Exploring Sequence-to-Sequence Learning in Aspect Term Extraction [paper] [code]

  • [ACL-2020] Don’t Eclipse Your Arts Due to Small Discrepancies: Boundary Repositioning with a Pointer Network for Aspect Extraction [paper]

OTE

  • [NAACL-2019] Target-oriented Opinion Words Extraction with Target-fused Neural Sequence Labeling [paper] [code]

AOTE

  • [AAAI-2017] Coupled Multi-Layer Attentions for Co-Extraction of Aspect and Opinion Terms [paper]

Emotion Classification

  • [EMNLP-2018] Improving Multi-label Emotion Classification via Sentiment Classification with Dual Attention Transfer Network [paper]

Projects & Competition

DataSet

Chinese

  • AI-Challenge [data]
  • SemEval ABSA 2016 [data]

English

  • Amazon product data [data]
  • Web data: Amazon reviews [data]
  • Amazon Fine Food Reviews [kaggle]
  • SemEval ABSA

Reference