Interactive Attention Networks for Aspect-Level Sentiment Classification

By Dehong Ma, Sujian Li, Xiaodong Zhang, Houfeng Wang. International Joint Conference on Artificial Intelligence, 2017

1,096 citations recorded by the source corpus.

Aspect-level sentiment classification aims at identifying the sentiment polarity of specific target in its context. Previous approaches have realized the importance of targets in sentiment classification and developed various methods with the goal of precisely modeling their contexts via generating target-specific representations. However, these studies always ignore the separate modeling of targets. In this paper, we argue that both targets and contexts deserve special treatment and need to be learned their own representations via interactive learning. Then, we propose the interactive attention networks (IAN) to interactively learn attentions in the contexts and targets, and generate the representations for targets and contexts separately. With this design, the IAN model can well represent a target and its collocative context, which is helpful to sentiment classification. Experimental results on SemEval 2014 Datasets demonstrate the effectiveness of our model.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/5e40be36d11483923cb12260bed6e8f7ed355ff3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Bing Liu
2. B. Liu
3. Hu Xu
4. Lei Shu
5. Lidong Bing
6. Erik Cambria
7. Sahisnu Mazumder
8. Shuai Wang

Citations in the corpus

• A Challenge Dataset and Effective Models for Aspect-Based Sentiment Analysis (2019, methodology): “…models: TD LSTM (Tang et al., 2016a), AT LSTM (Wang et al., 2016), ATAE LSTM (Wang et al., 2016), BiLSTM enhanced with attention mechanism, IAN (Ma et al., 2017) and AOA LSTM (Huang et al., 2018); (2) CNN based model: GCAE (Xue and Li, 2018); (3) Attention based models: MemNet (Tang et al.,…”
• Aspect Level Sentiment Classification with Attention-over-Attention Neural Networks (2018, methodology): “Along these lines, Ma et al. use two LSTM networks to model sentences and aspects separately [11].”
• Aspect-based Sentiment Classification with Aspect-specific Graph Convolutional Networks (2019, background): “, 2015) starts to play a critical role in more recent models (Wang et al., 2016; Tang et al., 2016b; Yang et al., 2017; Liu and Zhang, 2017; Ma et al., 2017; Huang et al., 2018).”
• Syntax-Aware Aspect Level Sentiment Classification with Graph Attention Networks (2019, methodology): “Based on these two attention vectors, it outputs a sentence representation and a target representation for classification (Ma et al., 2017).”
• Target-Dependent Sentiment Classification With BERT (2019, methodology): “TWO VARIANTS Sun et al. [36] construct an auxiliary sentence from the aspect and convert ABSA into a sentence-pair classification task.”
• Dependency Graph Enhanced Dual-transformer Structure for Aspect-based Sentiment Classification (2020, background): “, 2018), IAN (Ma et al., 2017), TNetLF (Li et al.”
• A General Survey on Attention Mechanisms in Deep Learning (2022, methodology): “Instead of using a sequential mechanism like alternating co-attention, [34] proposes the interactive co-attention mechanism that can calculate attention on both feature matrices in parallel, as depicted in Fig.”
• Dual Graph Convolutional Networks for Aspect-based Sentiment Analysis (2021, background): “To judge the sentiment of a particular aspect, previous studies (Wang et al., 2016; Tang et al., 2016a; Ma et al., 2017; Chen et al., 2017; Fan et al., 2018; Huang et al., 2018; Gu et al., 2018) have proposed various recurrent neural networks (RNNs) with attention mechanisms to generate aspect-speci”
• Relational Graph Attention Network for Aspect-based Sentiment Analysis (2020, background): “Most recent efforts (Wang et al., 2016b; Li et al., 2017; Ma et al., 2017; Fan et al., 2018) resort to assorted attention mechanisms to achieve this goal and have reported appealing results.”
• Deep Learning for Aspect-Based Sentiment Analysis: A Comparative Review (2019, background): “Ma et al. (2018) Reviews SemEval’15 modified (no explicit aspects) English LSTM + Target attention + Sentencelevel attention + SenticNet Macro F1: 76.”
• Transformation Networks for Target-Oriented Sentiment Classification (2018, methodology): “Ma et al. (2017) attempted to overcome this issue by measuring the importance score between each target word representation and the averaged sentence vector.”
• Attentional Encoder Network for Targeted Sentiment Classification (2019, background): “• IAN (Ma et al., 2017) learns the representations of the target and context with two LSTMs and attentions interactively, which generates the representations for targets and contexts with respect to each other.”
• Multi-grained Attention Network for Aspect-Level Sentiment Classification (2018, methodology): “MGAN-C only employs the coarse-grained attentions for prediction, which is similar with IAN.”
• Lifelong Learning Memory Networks for Aspect Sentiment Classification (2018, background): “There are other related studies [4], [21]–[23] using memory networks or the attention mechanism but with different focuses.”
• Transfer Capsule Network for Aspect Level Sentiment Classification (2019, methodology): “Settings We use Glove vectors with 840B to-kens (Pennington et al., 2014) Compared Methods To demonstrate the superiority of our TransCap for ASC tasks, we compare it with followings baselines: ATAE-LSTM (Wang et al., 2016b), IAN (Ma et al., 2017), AF-LSTM(CONV) (Tay et al., 2018), AF-LSTM(CORR) (Ta”
• Attention in Natural Language Processing (2019, methodology): “[111], an average (avg) is initially computed on each input and then used as a query in the application of attention to generate the embedding of the other input.”