Implicit Sentiment Analysis Based on Chain of Thought Prompting
Zhihua Duan, Jialin Wang
TL;DR
The paper tackles implicit sentiment analysis (ISA) by leveraging chain-of-thought prompting through the SAoT framework, which analyzes implicit aspects using common-sense reasoning, then reflects on the inference to deduce sentiment polarity. It evaluates the approach in zero-shot settings on SemEval-2014 Restaurant and Laptop datasets, demonstrating substantial improvements in F1 and ISA metrics when combined with ERNIE-Bot-4 (e.g., F1 of 75.27/76.50 and ISA of 66.29/73.46, respectively). The results illustrate that thought-inspired prompting can significantly enhance implicit sentiment understanding and prompt design for large language models. Overall, SAoT achieves strong zero-shot performance and reveals the practical potential of structured reasoning in sentiment analysis.
Abstract
Implicit Sentiment Analysis (ISA) is a crucial research area in natural language processing. Inspired by the idea of large language model Chain of Thought (CoT), this paper introduces a Sentiment Analysis of Thinking (SAoT) framework. The framework first analyzes the implicit aspects and opinions in the text using common sense and thinking chain capabilities. Then, it reflects on the process of implicit sentiment analysis and finally deduces the polarity of sentiment. The model is evaluated on the SemEval 2014 dataset, consisting of 1120 restaurant reviews and 638 laptop reviews. The experimental results demonstrate that the utilization of the ERNIE-Bot-4+SAoT model yields a notable performance improvement. Specifically, on the restaurant dataset, the F1 score reaches 75.27, accompanied by an ISA score of 66.29. Similarly, on the computer dataset, the F1 score achieves 76.50, while the ISA score amounts to 73.46. Comparatively, the ERNIE-Bot-4+SAoT model surpasses the BERTAsp + SCAPt baseline by an average margin of 47.99%.