Abstract:In the real world, multimodal sentiment analysis (MSA) enables the capture and analysis of sentiments by fusing multimodal information, thereby enhancing the understanding of real-world environments. The key challenges lie in handling the noise in the acquired data and achieving effective multimodal fusion. When processing the noise in data, existing methods utilize the combination of multimodal features to mitigate errors in sentiment word recognition caused by the performance limitations of automatic speech recognition (ASR) models. However, there still remains the problem of how to more efficiently utilize and combine different modalities to address the data noise. In multimodal fusion, most existing fusion methods have limited adaptability to the feature differences between modalities, making it difficult to capture the potential complex nonlinear interactions that may exist between modalities. To overcome the aforementioned issues, this paper proposes a new framework named multimodal-word-refinement and cross-modal-hierarchy (MWRCMH) fusion. Specifically, we utilized a multimodal word correction module to reduce sentiment word recognition errors caused by ASR. During multimodal fusion, we designed a cross-modal hierarchical fusion module that employed cross-modal attention mechanisms to fuse features between pairs of modalities, resulting in fused bimodal-feature information. Then, the obtained bimodal information and the unimodal information were fused through the nonlinear layer to obtain the final multimodal sentiment feature information. Experimental results on the MOSI-SpeechBrain, MOSI-IBM, and MOSI-iFlytek datasets demonstrated that the proposed approach outperformed other comparative methods, achieving Has0-F1 scores of 76.43%, 80.15%, and 81.93%, respectively. Our approach exhibited better performance, as compared to multiple baselines.

GLoMo: Global-Local Modal Fusion for Multimodal Sentiment Analysis

Global Local Fusion Neural Network for Multimodal Sentiment Analysis

Multimodal Sentiment Analysis Based on Cross-Modal Attention and Gated Cyclic Hierarchical Fusion Networks

Multi-level Multi-task Representation Learning with Adaptive Fusion for Multimodal Sentiment Analysis

Locally Confined Modality Fusion Network with a Global Perspective for Multimodal Human Affective Computing

Multimodal Sentiment Analysis with Word-Level Fusion and Reinforcement Learning

MMGLOTS: Multi-Modal Global-Local Transformer Segmentor for Remote Sensing Image Segmentation

Multimodal Sentiment Analysis in Realistic Environments Based on Cross-Modal Hierarchical Fusion Network

Multimodal Language Analysis with Recurrent Multistage Fusion

AdaMoW: Multimodal Sentiment Analysis Based on Adaptive Modality-Specific Weight Fusion Network

Improving Multimodal Fusion with Hierarchical Mutual Information Maximization for Multimodal Sentiment Analysis

VLP2MSA: Expanding Vision-Language Pre-Training to Multimodal Sentiment Analysis

Multimodal Mutual Attention-Based Sentiment Analysis Framework Adapted to Complicated Contexts

A cross modal hierarchical fusion multimodal sentiment analysis method based on multi-task learning

Two-Level Multimodal Fusion for Sentiment Analysis in Public Security

Multi-layer cross-modality attention fusion network for multimodal sentiment analysis

Multimodal Multi-loss Fusion Network for Sentiment Analysis

Multimodal Sentiment Analysis Network Based on Distributional Transformation and Gated Cross-Modal Fusion

A Graph Convolution-Based Heterogeneous Fusion Network for Multimodal Sentiment Analysis

Multi-Channel Attentive Graph Convolutional Network with Sentiment Fusion for Multimodal Sentiment Analysis

Tri-Modalities Fusion for Multimodal Sentiment Analysis