Syntax-Aware Graph Attention Network for Aspect-Level Sentiment Classification

Huang, Lianzhe; Sun, Xin; Li, Sujian; Zhang, Linhao; Wang, Houfeng

doi:10.18653/v1/2020.coling-main.69

Cited by 62 publications

(40 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Graph-based Methods. A number of recent works employ graph neural networks such as graph convolutional netowrks (GCN) [4], [6], [6], [9], [19] and graph attention networks (GAT) [3], [7], [20] to encode the syntax graph predicted by off-the-shelf dependency parsers. [1] employ a GCN to capture syntactic features and facilitate the information exchange between the aspect and its related context words.…”

Section: Related Workmentioning

confidence: 99%

DigNet: Digging Clues from Local-Global Interactive Graph for Aspect-level Sentiment Classification

Xing¹,

Tsang²

2022

Preprint

View full text Add to dashboard Cite

In aspect-level sentiment classification (ASC), state-of-the-art models encode either syntax graph or relation graph to capture the local syntactic information or global relational information. Despite the advantages of syntax and relation graphs, they have respective shortages which are neglected, limiting the representation power in the graph modeling process. To resolve their limitations, we design a novel local-global interactive graph, which marries their advantages by stitching the two graphs via interactive edges. To model this local-global interactive graph, we propose a novel neural network termed DigNet, whose core module is the stacked local-global interactive (LGI) layers performing two processes: intra-graph message passing and cross-graph message passing. In this way, the local syntactic and global relational information can be reconciled as a whole in understanding the aspect-level sentiment. Concretely, we design two variants of local-global interactive graphs with different kinds of interactive edges and three variants of LGI layers. We conduct experiments on several public benchmark datasets and the results show that we outperform previous best scores by 3%, 2.32%, and 6.33% in terms of Macro-F1 on Lap14, Res14, and Res15 datasets, respectively, confirming the effectiveness and superiority of the proposed local-global interactive graph and DigNet.

show abstract

Section: Related Workmentioning

confidence: 99%

DigNet: Digging Clues from Local-Global Interactive Graph for Aspect-level Sentiment Classification

Xing¹,

Tsang²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Graph Neural Networks (GNNs) have been used to address many problems that are inherently graph-like such as traffic networks, social networks, and physical and biological systems (Liu and Zhou, 2020). GNNs achieve impressive performance in many domains, including social networks (Wu et al, 2020) and natural science (Sanchez-Gonzalez et al, 2018) as well as NLP tasks like sentence classification (Huang et al, 2020), question generation (Pan et al, 2020), summarization (Fernandes et al, 2019) and derivational morphology (Hofmann et al, 2020). 7 github.com/robertostling/eflomal…”

Section: Annotation Projectionmentioning

confidence: 99%

Graph Neural Networks for Multiparallel Word Alignment

Imani,

Şenel,

Sabet

et al. 2022

Preprint

View full text Add to dashboard Cite

After a period of decrease, interest in word alignments is increasing again for their usefulness in domains such as typological research, cross-lingual annotation projection and machine translation. Generally, alignment algorithms only use bitext and do not make use of the fact that many parallel corpora are multiparallel. Here, we compute high-quality word alignments between multiple language pairs by considering all language pairs together. First, we create a multiparallel word alignment graph, joining all bilingual word alignment pairs in one graph. Next, we use graph neural networks (GNNs) to exploit the graph structure. Our GNN approach (i) utilizes information about the meaning, position and language of the input words, (ii) incorporates information from multiple parallel sentences, (iii) adds and removes edges from the initial alignments, and (iv) yields a prediction model that can generalize beyond the training sentences. We show that community detection provides valuable information for multiparallel word alignment. Our method outperforms previous work on three word alignment datasets and on a downstream task.

show abstract

“…• SAGAT [14] utilizes graph attention network and BERT to fully obtain both syntax and semantic information. • KGCapsAN-BERT [48] utilizes multi-prior knowledge to guide the capsule attention process and use a GCN-based syntactic layer to integrate the syntactic knowledge.…”

Section: Compared Baselinesmentioning

confidence: 99%

“…E-mail: Bowen.Xing@student.uts.edu.au, Ivor.Tsang@uts.edu.au between the aspect and its related words that are distant in context sequence. Graph Convolutional Networks (GCN) [12], [13] and Graph Attention Networks (GAT) [14], [15] are adopted to encode the syntax graphs predicted by off-the-shelf dependency parsers. [12], [16] employed GCNs to capture the local syntactic information.…”

Section: Introductionmentioning

confidence: 99%

Understand me, if you refer to Aspect Knowledge: Knowledge-aware Gated Recurrent Memory Network

Xing¹,

Tsang²

2021

Preprint

View full text Add to dashboard Cite

Aspect-level sentiment classification (ASC) aims to predict the fine-grained sentiment polarity towards a given aspect mentioned in a review. Despite recent advances in ASC, enabling machines to preciously infer aspect sentiments is still challenging. This paper tackles two challenges in ASC: (1) due to lack of aspect knowledge, aspect representation derived in prior works is inadequate to represent aspect's exact meaning and property information; (2) prior works only capture either local syntactic information or global relational information, thus missing either one of them leads to insufficient syntactic information. To tackle these challenges, we propose a novel ASC model which not only end-to-end embeds and leverages aspect knowledge but also marries the two kinds of syntactic information and lets them compensate for each other. Our model includes three key components: (1) a knowledge-aware gated recurrent memory network recurrently integrates dynamically summarized aspect knowledge; (2) a dual syntax graph network combines both kinds of syntactic information to comprehensively capture sufficient syntactic information; (3) a knowledge integrating gate re-enhances the final representation with further needed aspect knowledge; (4) an aspect-to-context attention mechanism aggregates the aspect-related semantics from all hidden states into the final representation. Experimental results on several benchmark datasets demonstrate the effectiveness of our model, which overpass previous state-of-the-art models by large margins in terms of both Accuracy and Macro-F1.

show abstract

Syntax-Aware Graph Attention Network for Aspect-Level Sentiment Classification

Cited by 62 publications

References 28 publications

DigNet: Digging Clues from Local-Global Interactive Graph for Aspect-level Sentiment Classification

DigNet: Digging Clues from Local-Global Interactive Graph for Aspect-level Sentiment Classification

Graph Neural Networks for Multiparallel Word Alignment

Understand me, if you refer to Aspect Knowledge: Knowledge-aware Gated Recurrent Memory Network

Contact Info

Product

Resources

About