摘要
随着脑科学与类脑智能研究的推进,基于脑电(Electroencephalogram,EEG)的情绪识别成为情感计算中的研究热点。近年来,深度学习(Deep Learning,DL)因其强大的特征提取与建模能力,在EEG情绪识别中展现出较高的预测精度和效率。系统回顾了DL在脑电情绪识别的研究进展,特别聚集于数据预处理、特征提取、分类模型的设计,并提出了“受试者依赖型”与“受试者独立型”双范式分类框架,为研究人员提供了清晰的研究路径。此外,还利用CiteSpace工具探讨了DL在脑电情绪识别领域的研究现状,从关键词共现、突现检测与聚类演化等角度综合分析,不仅为未来研究者提供参考,也为促进情感计算技术的实际应用奠定了基础。
With the advancement of neuroscience and brain-inspired intelligence research,EEG-based emotion recognition has become a hot topic in affective computing.In recent years,deep learning(DL),with its powerful feature extraction and modeling capabilities,has demonstrated high prediction accuracy and efficiency in EEG emotion recognition.This paper systematically reviews the research progress of DL in EEG emotion recognition,focusing particularly on data preprocessing,feature extraction,and classification model design.It also proposes a dual-paradigm classification framework of“subject-dependent”and“subject-independent”to provide researchers with clear research paths.Additionally,the paper utilizes the CiteSpace tool to explore the current state of DL research in EEG emotion recognition,conducting a comprehensive analysis from perspectives such as keyword co-occurrence,burst detection,and clustering evolution.This not only provides references for future researchers but also lays a foundation for promoting the practical application of affective computing technologies.
作者
李润依
张勇斌
付秀丽
LI Runyi;ZHANG Yongbin;FU Xiuli(School of Mechanical and Electrical Engineering,Beijing Institute of Graphic Communication,Beijing 102600,China;School of Information Engineering,Beijing Institute of Petrochemical Technology,Beijing 102617,China)
出处
《北京印刷学院学报》
2025年第12期70-78,共9页
Journal of Beijing Institute of Graphic Communication
基金
国家重点研发计划(2019YFB1707202)研究成果。
关键词
脑电图
情绪识别
深度学习
受试者独立
CITESPACE
electro encephalo gram(EEG)
emotion recognition
deep learning
subject-independent
CiteSpace
