Answers are grounded in this paper and cite their sources.
Systematic review reveals advancements in multimodal sentiment analysis, highlighting deep learning and fusion techniques.
Multimodal sentiment analysis (MSA) has emerged as an effective approach for assessing human emotions through several modalities, particularly verbal and non-verbal forms. Unimodal methods often fail to adequately capture the nuances of human affective states, whereas multimodal frameworks offer richer insight into sentiment by leveraging complementary sources of information. This study provides an overview of the MSA, fusion methods, and cutting-edge deep learning methods used in the development of MSA. It also discusses the datasets available and high-level deep learning models that have been implemented and, based on existing issues, provides future scope on developing MSA systems that are robust and scalable for real-world use cases.
No takes yet. Share an insight, caveat, or question.
Bargavi et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: