Recurrent Attention Network on Memory for Aspect Sentiment Analysis

By Peng Chen, Zhongqian Sun, Lidong Bing, Wei Yang. Conference on Empirical Methods in Natural Language Processing, 2017

913 citations recorded by the source corpus.

We propose a novel framework based on neural networks to identify the sentiment of opinion targets in a comment/review. Our framework adopts multiple-attention mechanism to capture sentiment features separated by a long distance, so that it is more robust against irrelevant information. The results of multiple attentions are non-linearly combined with a recurrent neural network, which strengthens the expressive power of our model for handling more complications. The weighted-memory mechanism not only helps us avoid the labor-intensive feature engineering work, but also provides a tailor-made memory for different opinion targets of a sentence. We examine the merit of our model on four datasets: two are from SemEval2014, i.e. reviews of restaurants and laptops; a twitter dataset, for testing its performance on social media data; and a Chinese news comment dataset, for testing its language sensitivity. The experimental results show that our model consistently outperforms the state-of-the-art methods on different types of data.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/8e3a3ba309fe68f5b1d02266a03395332f7d945c

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. Lidong Bing
5. Lei Shu
6. Erik Cambria
7. Sahisnu Mazumder
8. Shuai Wang

Citations in the corpus

• Exploiting BERT for End-to-End Aspect-based Sentiment Analysis (2019, background): “the aspectlevel sentiment classification) (Tang et al., 2016; Wang et al., 2016; Chen et al., 2017; Liu and Zhang, 2017; Ma et al., 2017, 2018b; Majumder et al., 2018; Li et al., 2018; He et al., 2018; Xue and Li, 2018; Wang et al., 2018; Fan et al., 2018; Huang and Carley, 2018; Lei et al., 2019; L”
• A Systematic Review on Affective Computing: Emotion Models, Databases, and Recent Advances (2022, methodology): “To decrease the training time and complexity based on LSTM and attention mechanisms to predict the sentiment polarity, Xue et al. [151] designed a Gated Convolutional network with Aspect Embedding (GCAE) for ACSA and ATSA, which are more efficient and simpler than the RNN-based model.”
• Deep Learning for Aspect-Based Sentiment Analysis: A Comparative Review (2019, methodology): “Thus, it can be inferred that CRFs can take advantage of the entire sentence sequence to estimate probability for the sentence labelling making CRF a frequent final classification layer of bidirectional RNNs (T. Chen et al., 2017; Irsoy & Cardie, 2014; Lample et al., 2016; P. Liu et al., 2015).”
• Lifelong Learning Memory Networks for Aspect Sentiment Classification (2018, background): “[21] considered learning an additional set of attention for aspect words, [22] suggested a recurrent attention mechanism, [23] differentiated attention from left and right context, and [4] provided solutions for target-sensitive sentiment.”
• A Unified Generative Framework for Aspect-based Sentiment Analysis (2021, neutral): “…model structures such as convolutional neural network (CNN) (Li et al., 2018a; Xue and Li, 2018), gated neural network (Zhang et al., 2016; Xue and Li, 2018), memory neural 2It is also referred to as the AE-OE co-Extraction. network (Tang et al., 2016b; Chen et al., 2017) have also been applied.”
• Sentiment Classification by Leveraging the Shared Knowledge from a Sequence of Domains (2019, methodology): “Many aspect-based sentiment analysis methods are also reported [5,14,17,26,33,36,37,39].”
• Attentional Encoder Network for Targeted Sentiment Classification (2019, background): “s to build higher semantic information. Wang et al. (2016) propose ATAE-LSTM which concatenates target embeddings with word representations and let targets participate in computing attention weights. Chen et al. (2017) propose RAM which adopts multiple-attention mechanism on the memory built with bi”
• Aspect Based Sentiment Analysis with Gated Convolutional Networks (2018, methodology): “We classify the existing work into two main categories based on the descriptions of sentiment analysis tasks in SemEval 2014 Task 4 (Pontiki et al., 2014): Aspect-Term Sentiment Analysis and Aspect-Category Sentiment Analysis.”
• Transformation Networks for Target-Oriented Sentiment Classification (2018, methodology): “To overcome this limitation, researchers bring in the attention mechanism to model target-context association (Tang et al., 2016a,b; Wang et al., 2016; Yang et al., 2017; Liu and Zhang, 2017; Ma et al., 2017; Chen et al., 2017; Zhang et al., 2017; Tay et al., 2017).”
• 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”
• Position-Aware Tagging for Aspect Sentiment Triplet Extraction (2020, background): “Such a task either refers to predicting sentiment polarity for a given target (Dong et al., 2014; Chen et al., 2017; Xue and Li, 2018; Wang and Lu, 2018; Wang et al., 2018; Li et al., 2018a; Peng et al., 2018; Xu et al., 2020) or joint extraction of targets as well as sentiment associated with each ”
• Aspect-based sentiment analysis via affective knowledge enhanced graph convolutional networks (2021, neutral): “[14] explored a multiple attention-based memory network to learn the long distance sentiment features.”
• Aspect-based Sentiment Classification with Aspect-specific Graph Convolutional Networks (2019, methodology): “based sentiment classification are mainly based on neural network methods (Dong et al., 2014; Vo and Zhang, 2015). Ever since Tang et al. (2016a) pointed out the challenge of modelling semantic relatedness between context words and aspects, attention mechanism coupled with Recurrent Neural Networks ”
• A Survey on Text Classification: From Shallow to Deep Learning (2020, neutral): “ 91.3 - - - - - - - BLSTM-2DCNN [118] 82.3 89.5 - - - - 96.5 - - - TopicRNN [57] - - 93.72 - - - - - - - DPCNN [59] - - - 97.36 69.42 65.19 - 93.13 99.12 - KPCNN [121] 83.25 - - - - - - 88.36 - - RAM [65] - - - - - - - - - - RNN-Capsule [119] 83.8 - - - - - - - - ULMFiT [70] - - 95.4 97.84 71.02 - -”