Downloads: 1
Analysis Study Research Paper | Data and Knowledge Engineering | Volume 15 Issue 3, March 2026 | Pages: 1612 - 1615 | India
Multimodal Sentiment Analysis: A Systematic Review
Abstract: Multimodal aspect-based sentiment analysis has emerged as a crucial research area, integrating information from diverse modalities like text, images, audio, and video to analyze human emotions and sentiments at the aspect level. This review synthesizes recent advancements in multimodal aspect-based sentiment analysis, highlighting innovative approaches to fusing text, images, and other modalities to gain deeper insights into human emotions. We examine the strengths and limitations of various methodologies, including interactive memory networks and cross-modal attention mechanisms, and discuss the challenges of bridging the semantic gap between modalities. Furthermore, we identify key areas for future research, including the need for standardized evaluation frameworks and the potential benefits of incorporating additional modalities. By providing a comprehensive overview of this rapidly evolving field, this review aims to inform and inspire future research in multimodal aspect-based sentiment analysis.
Keywords: Sentiment Analysis, NLP (Natural Language Processing), Multimodal Fusion, Cross-Modal Attention, Interactive Memory Networks
How to Cite?: A. Martina Betsy, Dr. Sumathy Kingslin, "Multimodal Sentiment Analysis: A Systematic Review", Volume 15 Issue 3, March 2026, International Journal of Science and Research (IJSR), Pages: 1612-1615, https://www.ijsr.net/getabstract.php?paperid=SR26327075929, DOI: https://dx.dx.doi.org/10.21275/SR26327075929