A method for analyzing emotional change trend in an interactive situation and application thereof
By collecting multi-source heterogeneous data, using adaptive time window fusion, and employing nonlinear dynamic models, the problems of lagging emotion prediction and unclear attribution in existing technologies have been solved. This enables accurate prediction and proactive intervention of emotion change trends, improving the human-computer interaction experience and user emotional health.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHEJIANG QIANYING BIRDHOUSE FORESTRY DEV CO LTD
- Filing Date
- 2026-04-16
- Publication Date
- 2026-07-10
AI Technical Summary
Existing technologies struggle to accurately predict the critical point of emotional evolution during long-term interactions and lack the ability to attribute situational causes to causality, making it impossible to proactively intervene before emotions spiral out of control.
By collecting multi-source heterogeneous data in real time, performing edge privacy desensitization processing and adaptive multi-scale time window fusion, analyzing the trajectory of emotion evolution using a nonlinear dynamic model, constructing a context-emotion dynamic causal graph, generating emotion attribution reports, and implementing targeted interactive intervention strategies.
It enables accurate prediction and proactive intervention of emotion change trends, improves human-computer interaction experience and user emotional health, and solves the problems of lagging emotion prediction and unclear attribution in traditional technologies.
Smart Images

Figure CN122364762A_ABST