Deep learning for sentiment analysis: A survey

By Lei Zhang, Shuai Wang, B. Liu. WIREs Data Mining Knowl. Discov., 2018

1,916 citations recorded by the source corpus.

Deep learning has emerged as a powerful machine learning technique that learns multiple layers of representations or features of the data and produces state‐of‐the‐art prediction results. Along with the success of deep learning in many application domains, deep learning is also used in sentiment analysis in recent years. This paper gives an overview of deep learning and then provides a comprehensive survey of its current applications in sentiment analysis.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/2b91a2cbcd9cce902cbc8da78fec5f18f4bffc98

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

Bi-LSTM Model to Increase Accuracy in Text Classification: Combining Word2vec CNN and Attention Mechanism (2020, background): “Natural-language-based sentiment classiﬁcation has a wide range of applications, such as movie review classiﬁcation, subjective and objective sentence classiﬁcation, and text classiﬁcation technology [1].”
Sentiment Analysis for E-Commerce Product Reviews in Chinese Based on Sentiment Lexicon and Deep Learning (2020, methodology): “The deep learning models commonly used in the ﬁeld of text sentiment analysis are CNN, Recurrent Neural Network (RNN), LSTM, and Gated Recurrent Unit (GRU) [22].”
Sentiment analysis on massive open online course evaluations: A text mining and deep learning approach (2020, methodology): “Based on all the earlier computations, the output has been determined [73].”
Leveraging Just a Few Keywords for Fine-Grained Aspect Detection Through Weakly Supervised Co-Training (2019, positive): “Rule-based or traditional learning models for aspect detection have been outperformed by supervised neural networks (Liu et al., 2015; Poria et al., 2016; Zhang et al., 2018).”
Adapting BERT for Target-Oriented Multimodal Sentiment Classification (2019, methodology): “) has recently been leveraged to provide complementary sentiment signals to traditional textual features [Zhang et al., 2018a].”
Lifelong Learning Memory Networks for Aspect Sentiment Classification (2018, background): “Its key advantages to ASC are its ability to learn aspect and context representation (in an embedding manner) and its attention mechanism [10], [11].”
Explainable Sentiment Analysis with Applications in Medicine (2018, background): “In order to perform sentiment analysis with a massive amount of data, as for example social media data, deep learning algorithms have recently gained a lot a popularity, by achieving state-of-the-art results in various tasks [1].”
Exploring AI Ethics of ChatGPT: A Diagnostic Analysis (2023, background): “models to make informed decisions based on natural language input, as demonstrated in studies on sentiment analysis [36], text classification [37], and question answering [38].”
Deep Learning for Aspect-Based Sentiment Analysis: A Comparative Review (2019, methodology): “The main strength of the CNN is its ability to extract the most important n-gram features from the input to create an "informative latent semantic representation" for undertaking further classification tasks (Rojas-Barahona, 2016; Young et al., 2017; L. Zhang et al., 2018).”
Speed Reading: Learning to Read ForBackward via Shuttle (2018, methodology): “…processing tasks, e.g., part-of-speech (POS) tagging (Wang et al., 2015), named-entity recognition (Chiu and Nichols, 2015), sentiment analysis (Zhang et al., 2018), document classification (Kim, 2014; Le and Mikolov, 2014a), cloze (Srinivasan et al., 2018), machine translation (Bahdanau et…”
FPGA-Based Accelerators of Deep Learning Networks for Learning and Classification: A Review (2019, background): “The field of DL emerged in 2006 after a long pause in the area of neural networks (NNs) research [4].”
Vocabulary Learning via Optimal Transport for Neural Machine Translation (2020, positive): “Due to the discreteness of text, vocabulary construction ( vocabularization for short) is a prerequisite for neural machine translation (NMT) and many other natural language processing (NLP) tasks using neural networks (Mikolov et al., 2013; Vaswani et al., 2017; Gehrmann et al., 2018; Zhang et al.,”
Target-Dependent Sentiment Classification With BERT (2019, background): “RELATED WORK As shown in [48], there are three important tasks in ABSA.”
Sentiment Analysis Based on Deep Learning: A Comparative Study (2020, methodology): “It is a deep neural network architecture [22], typically composed of convolutional and pooling or subsampling layers to provide inputs to a fully-connected classification layer.”
A Review of Machine Learning and Deep Learning Applications (2018, background): “It means that deep learning is suited for analyzing and extracting useful knowledge from both large huge amounts of data and data collected from different sources [14].”