Semantic and Syntactic Enhanced Aspect Sentiment Triplet Extraction

Zhexue Chen,Hai Jin,Hong Huang,Bang Liu,Xuanhua Shi

doi:10.18653/v1/2021.findings-acl.128

Abstract

Aspect Sentiment Triplet Extraction (ASTE) aims to extract triplets from sentences, where each triplet includes an entity, its associated sentiment, and the opinion span explaining the reason for the sentiment. Most existing research addresses this problem in a multi-stage pipeline manner, which neglects the mutual information between such three elements and has the problem of error propagation. In this paper, we propose a Semantic and Syntactic Enhanced aspect Sentiment triplet Extraction model (S3E2) to fully exploit the syntactic and semantic relationships between the triplet elements and jointly extract them. Specifically, we design a Graph-Sequence duel representation and modeling paradigm for the task of ASTE: we represent the semantic and syntactic relationships between word pairs in a sentence by graph and encode it by Graph Neural Networks (GNNs), as well as modeling the original sentence by LSTM to preserve the sequential information. Under this setting, we further apply a more efficient inference strategy for the extraction of triplets. Extensive evaluations on four benchmark datasets show that S3E2 significantly outperforms existing approaches, which proves our S3E2's superiority and flexibility in an end-to-end fashion.

Highlights

Aspect-based Sentiment Analysis (ABSA) usually requires to extract comment targets in a review and judge corresponding sentiment polarities (Liu, 2012; Pontiki et al, 2014)
When we compare S3E2 with competitive baselines, Grid-CNN and Grid-BiLSTM in detail, we find that the reason why we perform better on 14res and 15res is because we extract a more complete set of triplets in these two datasets, resulting a more significant recall
The reason why we perform better on 14lap and 16res is because we extract more accurate triplets, resulting a more significant precision. Such comprehensive results demonstrate the strength of S3E2, which has the ability to learn multi-faceted semantics and and is good at extracting triplets

Summary

Introduction

Aspect-based Sentiment Analysis (ABSA) usually requires to extract comment targets in a review and judge corresponding sentiment polarities (Liu, 2012; Pontiki et al, 2014). One naive solution is to split the ASTE task into two stages in a pipeline manner using a unified tagging schema 1 (Peng et al, 2020) Such a pipeline approach lacks an effective mechanism to capture the three elements’ relationship and suffers from error propagation. Another solution for the ASTE task is to use an end-to-end model to extract triplets (Xu et al, 2020; Wu et al, 2020). These methods focus on designing a new tagging schema to formalize ASTE into a unified task and cannot effectively establish the connection between words and ignore the semantic and syntactic relationship between the three elements

Objectives

Methods

Results

Conclusion