Teaching video sentiment analysis method and system based on multi-modal analysis
Through the teaching video sentiment analysis method based on multimodal analysis, multimodal data in teaching videos is extracted and integrated, and event recognition and sentiment analysis is solved, and the problem of insufficient integration of teacher emotions and multimodal data in the existing technology is solved, and more accurate and comprehensive sentiment analysis is achieved.
Patent Information
- Application Number
- CN202411723999.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-28
- Publication Date
- 2025-03-18
- Estimated Expiration
- 2044-11-28
AI Technical Summary
The existing teaching video sentiment analysis technology ignores teacher emotions, lacks integration of multimodal data, and the limitations of emotional event recognition models, making it difficult to fully demonstrate the synergistic relationship of multimodal data in different situations.
A teaching video emotion analysis method based on multimodal analysis is proposed. By obtaining classroom teaching videos, students' face information, teaching audio information and teaching text information are extracted, and a multimodal emotion analysis model is input to the multimodal emotion analysis model to identify events, and a multimodal teaching emotional interaction visual analysis system is built to conduct emotional evolution analysis, student group emotional correlation analysis and teacher-student teaching emotional interaction analysis.
Effectively pay attention to the emotional changes of teachers and students, integrate multimodal data, improve analysis accuracy in complex situations, provide intuitive and credible emotional analysis data, and support classroom effect evaluation.