Multi-modal sentiment analysis model and method, electronic equipment and medium

By using a multimodal sentiment analysis model and employing feature enhancement, multimodal balancing, and hierarchical fusion prediction modules, the problem of neglecting deep coupling mechanisms between modalities is solved, resulting in more accurate sentiment prediction and avoiding the influence of modal feature noise and redundant information.

CN121659191APending Publication Date: 2026-03-13ZHENGZHOU UNIVERSITY OF LIGHT INDUSTRY
0 Cites 1 Cited by

Patent Information

Application Number
CN202511708321.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-11-19
Publication Date
2026-03-13

AI Technical Summary

Technical Problem

Existing multimodal sentiment analysis methods rely on feature alignment between modalities, neglecting the deep coupling mechanism between modalities. This makes it difficult for traditional methods to fully capture complex emotions, and modal features are often noisy, with important semantics being buried by redundant information.

Method used

A multimodal sentiment analysis model is adopted. Multimodal feature representations are extracted through a feature enhancement module and enhanced using a graph convolutional network. A multimodal balancing module is combined for alignment and noise reduction decoupling. Generative adversarial networks and Transformer architecture are used to optimize the distribution of modal features. Finally, sentiment prediction is performed through a hierarchical fusion prediction module.

Benefits of technology

It effectively reduces the difference between structured and unstructured information between modalities, weakens modal noise, fully explores deep-seated specific information, avoids the limitations of traditional methods, and achieves more accurate sentiment prediction.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention provides a multi-modal sentiment analysis model and method, electronic equipment and a medium, and the model comprises a feature enhancement module which is used for extracting original multi-modal data features through an exclusive tool, constructing a graph structure, and enhancing the graph structure through a graph convolutional network to obtain multi-modal enhanced features; the modal multi-stage balance module is used for processing enhanced features by using different multi-stage network structures and outputting multi-modal consistency representation; the modal noise reduction decoupling and specificity recombination module is used for obtaining low-noise representation through a modal noise reduction decomposer based on the global information and recombining the low-noise representation in a modal bank to generate low-noise multi-modal specificity representation; and the hierarchical fusion prediction module is used for fusing consistency and specificity representation according to single-peak, double-peak and three-peak modes, and outputting an emotion prediction result through a multi-layer perceptron.
Need to check novelty before this filing date? Find Prior Art