SGPT: Semantic Graphs based Pre-training for Aspect-based Sentiment Analysis

TL;DR

This paper introduces SGPT, a pre-training approach that incorporates semantic graphs to enhance sentiment analysis by capturing synonym and aspect-sentiment relationships, leading to improved performance over existing models.

Contribution

The paper proposes a novel semantic graph-based pre-training method that explicitly encodes sentiment-related semantic information into language models for better sentiment analysis.

Findings

SGPT outperforms strong pre-trained baselines on multiple sentiment analysis tasks.

Semantic graphs effectively encode synonym and aspect-sentiment relations.

Enhanced models show improved understanding of sentiment nuances.

Abstract

Previous studies show effective of pre-trained language models for sentiment analysis. However, most of these studies ignore the importance of sentimental information for pre-trained models.Therefore, we fully investigate the sentimental information for pre-trained models and enhance pre-trained language models with semantic graphs for sentiment analysis.In particular, we introduce Semantic Graphs based Pre-training(SGPT) using semantic graphs to obtain synonym knowledge for aspect-sentiment pairs and similar aspect/sentiment terms.We then optimize the pre-trained language model with the semantic graphs.Empirical studies on several downstream tasks show that proposed model outperforms strong pre-trained baselines. The results also show the effectiveness of proposed semantic graphs for pre-trained model.