Abstract:The DimABSA task requires fine-grained sentiment intensity prediction for restaurant reviews, including scores for Valence and Arousal dimensions for each Aspect Term. In this study, we propose a Coarse-to-Fine In-context Learning(CFICL) method based on the Baichuan2-7B model for the DimABSA task in the SIGHAN 2024 workshop. Our method improves prediction accuracy through a two-stage optimization process. In the first stage, we use fixed in-context examples and prompt templates to enhance the model's sentiment recognition capability and provide initial predictions for the test data. In the second stage, we encode the Opinion field using BERT and select the most similar training data as new in-context examples based on similarity. These examples include the Opinion field and its scores, as well as related opinion words and their average scores. By filtering for sentiment polarity, we ensure that the examples are consistent with the test data. Our method significantly improves prediction accuracy and consistency by effectively utilizing training data and optimizing in-context examples, as validated by experimental results.

What problem does this paper attempt to address?

### Problems Addressed by the Paper The paper attempts to address the problem of **Dimensional Aspect-Based Sentiment Analysis (DimABSA)**, particularly in the context of restaurant reviews. Specifically, this task requires predicting the sentiment intensity for each aspect term in a review, including two dimensions: Valence and Arousal. ### Detailed Background - **Aspect-Based Sentiment Analysis (ABSA)** is an important research topic in Natural Language Processing (NLP), aiming to identify various aspects of a given entity and analyze the sentiment polarity associated with each aspect. - The **DimABSA** task further extends ABSA by not only identifying sentiment polarity but also predicting the continuous numerical intensity of the sentiment, providing more granular information in sentiment analysis. - **Valence** indicates the pleasantness of the sentiment, while **Arousal** indicates the level of excitement or calmness of the sentiment. - **Challenges**: Existing methods face issues of insufficient accuracy and poor consistency when dealing with multi-dimensional sentiment intensity prediction. ### Proposed Solution in the Paper - **Coarse-to-Fine In-context Learning (CFICL)**: - **First Stage**: Enhance the model's sentiment recognition capability using fixed context examples and prompt templates, and make preliminary predictions on the test data. - **Second Stage**: Utilize BERT to encode opinion fields, select the most similar training data as new context examples, and further optimize the prediction results. ### Main Contributions - **Improved Prediction Accuracy**: The two-stage method significantly improves the model's prediction accuracy in both Valence and Arousal dimensions. - **Reduced Sentiment Polarity Bias**: By carefully selecting and constructing context examples, the method effectively reduces sentiment polarity bias. - **Experimental Validation**: Experimental results show that the method performs excellently in multiple sub-tasks, especially achieving significant improvements in the sentiment quadruple extraction task. ### Conclusion The paper proposes a two-stage in-context learning method based on the Baichuan2-7B model to address the fine-grained sentiment intensity prediction problem in the DimABSA task. By gradually optimizing context examples, the method significantly improves prediction accuracy and consistency, providing valuable references for future fine-grained sentiment analysis models.

ZZU-NLP at SIGHAN-2024 dimABSA Task: Aspect-Based Sentiment Analysis with Coarse-to-Fine In-context Learning

Prior-Bert and Multi-Task Learning for Target-Aspect-Sentiment Joint Detection

Towards Unifying the Label Space for Aspect- and Sentence-based Sentiment Analysis

A Hybrid Approach to Dimensional Aspect-Based Sentiment Analysis Using BERT and Large Language Models

FaiMA: Feature-aware In-context Learning for Multi-domain Aspect-based Sentiment Analysis

A Local and Global Context Focus Multilingual Learning Model for Aspect-Based Sentiment Analysis

Fusion Local and Global Aspect-based Sentiment Analysis

Feature distillation network for aspect-based sentiment analysis

Aspect-Pair Supervised Contrastive Learning for aspect-based sentiment analysis

A syntactic features and interactive learning model for aspect-based sentiment analysis

Aspect-Based Sentiment Analysis using Local Context Focus Mechanism with DeBERTa

A Dependency Syntactic Knowledge Augmented Interactive Architecture for End-to-End Aspect-based Sentiment Analysis

Aspect-based Sentiment Analysis Via Fusing Multiple Sources of Textual Knowledge.

Modeling Inter-Aspect Dependencies with a Non-temporal Mechanism for Aspect-Based Sentiment Analysis.

A Contrastive Cross-Channel Data Augmentation Framework for Aspect-based Sentiment Analysis

Aspect-Level Sentiment Analysis Based on Deep Learning

Co-attention networks based on aspect and context for aspect-level sentiment analysis

Cross-Domain Aspect-based Sentiment Classification with Pre-Training and Fine-Tuning Strategy for Low-Resource Domains

Fine Granular Aspect Analysis Using Latent Structural Models.

Enhanced dual-level dependency parsing for aspect-based sentiment analysis

Aspects Are Anchors: Towards Multimodal Aspect-based Sentiment Analysis Via Aspect-driven Alignment and Refinement