Psychological sentiment analysis and human-computer interaction optimization model under multimodal data fusion

Multimodal data fusion (MDF) is a data processing technology in the field of artificial intelligence, which refers to the use of computers for the integrated processing of multimodal data, pictures, text, audio and video. Multimodal data fusion technology has important research and application value in psychological sentiment analysis and human-computer interaction optimization. Compared with traditional unimodal data processing, MDF has higher accuracy and reliability in analyzing the described objects. This study combines the fields of psychology and artificial intelligence, and proposes a multimodal data fusion-based sentiment analysis model, which realizes accurate recognition and classification of emotional states by integrating multimodal information, such as text, speech and images. The study designs a multimodal feature extraction and fusion algorithm, and optimizes the model performance by combining deep learning techniques. This paper explores the application scenarios of sentiment analysis in optimizing human-computer interaction, including intelligent assistants, educational technology and mental health management. The experimental results show that the model is better than traditional methods in terms of accuracy and stability of emotion recognition, providing theoretical basis and technical support for building a more natural and efficient emotional human-computer interaction system.