Lu, Qiang and Sun, Xia and Long, Yunfei and Gao, Zhizezhang and Feng, Jun and Sun, Tao (2023) Sentiment Analysis: Comprehensive Reviews, Recent Advances, and Open Challenges. IEEE Transactions on Neural Networks and Learning Systems, 35 (11). pp. 15092-15112. DOI https://doi.org/10.1109/TNNLS.2023.3294810
Lu, Qiang and Sun, Xia and Long, Yunfei and Gao, Zhizezhang and Feng, Jun and Sun, Tao (2023) Sentiment Analysis: Comprehensive Reviews, Recent Advances, and Open Challenges. IEEE Transactions on Neural Networks and Learning Systems, 35 (11). pp. 15092-15112. DOI https://doi.org/10.1109/TNNLS.2023.3294810
Lu, Qiang and Sun, Xia and Long, Yunfei and Gao, Zhizezhang and Feng, Jun and Sun, Tao (2023) Sentiment Analysis: Comprehensive Reviews, Recent Advances, and Open Challenges. IEEE Transactions on Neural Networks and Learning Systems, 35 (11). pp. 15092-15112. DOI https://doi.org/10.1109/TNNLS.2023.3294810
Abstract
Sentiment analysis aims to understand the attitudes and views of opinion holders with computers. Previous studies have achieved significant breakthroughs and extensive applications in the past decade, such as public opinion analysis and intelligent voice service. With the rapid development of deep learning, sentiment analysis based on various modalities has become a research hotspot. However, only individual modality has been analyzed separately, lacking a systematic carding of comprehensive sentiment analysis methods. Meanwhile, few surveys covering the topic of multi-modal sentiment analysis have been explored yet. In this paper, we first take the modality as the thread to design a novel framework of sentiment analysis tasks to provide researchers with a comprehensive understanding of relevant advances on sentiment analysis. Then, we introduce the general workflows and recent advances of single-modal in detail, discuss the similarities and differences of single-modal sentiment analysis in data processing and modeling to guide multi-modal sentiment analysis. And summarize the commonly used datasets to provide a guidance on data and methods for researchers that according to different task types. Next, a new taxonomy is proposed to fill the research gaps in multi-modal sentiment analysis, which is divided into multi-modal representation learning and multi-modal data fusion. The similarities and differences between these two methods and the latest advances are described in detail, such as dynamic interaction between multi-modalities, and the multi-modal fusion technologies are further expanded. Moreover, we explore the advanced studies on multi-modal alignment, chatbots and ChatGPT in sentiment analysis. Finally, we discuss the open research challenges of multimodal sentiment analysis, and provide four potential aspects to improve future works, such as cross-modal contrastive learning and multi-modal pre-training models.
Item Type: | Article |
---|---|
Uncontrolled Keywords: | Multimodal data fusion; multimodal representation learning; multimodal, sentiment analysis (SA); single-modal |
Divisions: | Faculty of Science and Health Faculty of Science and Health > Computer Science and Electronic Engineering, School of |
SWORD Depositor: | Unnamed user with email elements@essex.ac.uk |
Depositing User: | Unnamed user with email elements@essex.ac.uk |
Date Deposited: | 19 Sep 2023 16:01 |
Last Modified: | 02 Nov 2024 07:38 |
URI: | http://repository.essex.ac.uk/id/eprint/35942 |
Available files
Filename: Sentiment_Analysis_Comprehensive_Reviews_Recent_Advances_and_Open_Challenges.pdf