EEG融入自标签的自监督对比学习情感分析方法

Self-supervised Contrastive Learning Sentiment Analysis Method with EEG Incorporating Self-labeling

基于脑电信号的情绪识别是推断人类内心情绪状态的有效方法。近年来脑电信号的发展主要受标签标注的限制,如何从无标签脑电数据中学习到稳健的矢量表达是当前研究的重点。本文在自监督模型基础上,将经特征拓扑后的脑电数据作为输入,在训练过程中融入自标签Top-K,同时引入注意力机制,使模型获得更好的性能。模型在脑电数据集SEED上取得了78.3%的分类准确度。

Emotion recognition based on EEG signals is an effective method to infer the emotional state of human beings.In recent years,the development of EEG signals is mainly limited by labels.How to learn robust vector representations from unlabeled EEG data is the focus of current research.Based on the self-supervised model,this paper takes the EEG data after feature topology as input,incorporates the self-label Top-K in the training process,and introduces an attention mechanism to make the model achieve better performance.The model achieved a classification accuracy of 78.3%on the EEG dataset SEED.