Targeted Aspect-Based Sentiment Analysis via Embedding Commonsense Knowledge into an Attentive LSTM

By Yukun Ma, Haiyun Peng, Erik Cambria. 2018

583 citations recorded by the source corpus.

Analyzing people’s opinions and sentiments towards certain aspects is an important task of natural language understanding. In this paper, we propose a novel solution to targeted aspect-based sentiment analysis, which tackles the challenges of both aspect-based sentiment analysis and targeted sentiment analysis by exploiting commonsense knowledge. We augment the long short-term memory (LSTM) network with a hierarchical attention mechanism consisting of a target-level attention and a sentence-level attention. Commonsense knowledge of sentiment-related concepts is incorporated into the end-to-end training of a deep neural network for sentiment classification. In order to tightly integrate the commonsense knowledge into the recurrent encoder, we propose an extension of LSTM, termed Sentic LSTM. We conduct experiments on two publicly released datasets, which show that the combination of the proposed attention architecture and Sentic LSTM can outperform state-of-the-art methods in targeted aspect sentiment tasks.

Subfields: Artificial Intelligence

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

Citations in the corpus

Recent Trends in Deep Learning Based Natural Language Processing (2017, background): “[101] augmented LSTM with a hierarchical attention mechanism consisting of a target-level attention and a sentence-level attention to exploit com-”
Utilizing BERT for Aspect-Based Sentiment Analysis via Constructing Auxiliary Sentence (2019, methodology): “When evaluating the aspect detection, following Ma et al. (2018), we use strict accuracy and Macro-F1, and we also report AUC.”
Attention in Natural Language Processing (2019, background): “[113] identified a single “target” macro-object T as a set of words, which do not necessarily have to form a sequence in D, and then used such a macro-object as keys, KT .”
A survey on empathetic dialogue systems (2020, background): “A similar problem has been raised for emotion or sentiment classification [140, 141, 142], but has been missing in the context of dialogue systems.”
Recent Trends in Deep Learning Based Natural Language Processing [Review Article] (2018, background): “[101] augmented LSTM with a hierarchical attention mechanism consisting of a target-level attention and a sentence-level attention to exploit commonsense knowledge for targeted aspect-based sentiment analysis.”
Beneath the Tip of the Iceberg: Current Challenges and New Directions in Sentiment Analysis Research (2020, background): “onsense for associating aspects with their sentiments can be highly beneﬁcial for this task. Commonsense knowledge graphs connect the aspects to various sentiment-bearing concepts via semantic links (Ma et al., 2018). Additionally, semantic links between words can be utilized to mine associations be”
Deep Learning for Aspect-Based Sentiment Analysis: A Comparative Review (2019, background): “…(2016) Hotel DBS Text Mining Challenge 2015 data English Weighted Hierachi Bidirectional RNN (mini-batches) 65% (10 aspects); 55% (20 aspects) 18 Y. Ma et al. (2018) Reviews SemEval’15 modified (no explicit aspects) English LSTM + Target attention + Sentencelevel attention + SenticNet Macro F1:…”
Aspect-based sentiment analysis via affective knowledge enhanced graph convolutional networks (2021, unknown): “[50] incorporated commonsense knowledge into the LSTM model to extract target-level and 130”
Exploiting BERT for End-to-End Aspect-based Sentiment Analysis (2019, background): “…the second one and the third one, namely, Aspect-oriented Opinion Words Extraction (AOWE) (Fan et al., 2019) and End-to-End Aspect-based Sentiment Analysis (E2E-ABSA) (Ma et al., 2018a; Schmitt et al., 2018; Li et al., 2019a; Li and Lu, 2017, 2019), are related to a sequence tagging problem.”
Multimodal Sentiment Analysis using Hierarchical Fusion with Context Modeling (2018, background): “While most works approach it as a simple categorization problem, sentiment analysis is actually a suitcase research problem [16] that requires tackling many NLP tasks, including word polarity disambiguation [17], subjectivity detection [18], personality recognition [19], microtext normalization [20]”
A Survey on Application of Knowledge Graph (2020, background): “[54] improve the classic LSTM cell with adding components integrating with external knowledge, which directly contributes to the identification of aspects and sentiment polarity for sentiment analysis.”
Towards Generalizable Neuro-Symbolic Systems for Commonsense Question Answering (2019, methodology): “Ma et al. (2018) proposed to used embedded affective commonsense knowledge inside LSTM cell to control the information ﬂow in each gate for sentiment analysis task.”
Transfer Capsule Network for Aspect Level Sentiment Classification (2019, background): “In recent years, a number of deep learning methods have been proposed for aspect-level sentiment classiﬁcation (ASC) tasks (Dong et al., 2014; Vo and Zhang, 2015; Tang et al., 2016; Wang et al., 2016a; Ma et al., 2017; Li et al., 2018; Ma et al., 2018; Wang et al., 2018a).”