Scholarlink

<p>Automatic personality recognition (APR) is an emerging field of interdisciplinary research between psychology and computer science, aiming to infer individual’s personality from signals perceived by machines. This study introduces an innovative electroencephalogram (EEG) processing approach called LSSENet, which leverages both long short-term memory (LSTM) and squeeze-and-excitation network (SENet) to automatically classify the Big Five personality traits of people. LSSENet has the ability to simultaneously capture the spatial and temporal characteristics of EEG signals. As for the spatial dimension, LSSENet uses SENet to adaptively learn the contribution weights of different EEG channels. In terms of temporal information, LSTM is integrated into the model to identify the short-term coherence and long-term consistency features. To assess the effectiveness of LSSENet, a comprehensive set of experiments was carried on the AMIGOS dataset. The results show that LSSENet outperforms other EEG-based personality recognition methods, achieving the best classification average accuracy in every dimension of Big Five personality traits.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

EEG-based personality recognition with long short-term memory and squeeze-and-excitation network

  • 期刊论文

  • 发表日期：

    2025年10月15日

查看全文
  • Yuangang Wang,
  • Xu Chen,
  • Xueting Jiang,
  • Haoran Liu,
  • Shuo Guan,
  • Xiaodong Liu,
  • Xiaodong Duan

摘要

Automatic personality recognition (APR) is an emerging field of interdisciplinary research between psychology and computer science, aiming to infer individual’s personality from signals perceived by machines. This study introduces an innovative electroencephalogram (EEG) processing approach called LSSENet, which leverages both long short-term memory (LSTM) and squeeze-and-excitation network (SENet) to automatically classify the Big Five personality traits of people. LSSENet has the ability to simultaneously capture the spatial and temporal characteristics of EEG signals. As for the spatial dimension, LSSENet uses SENet to adaptively learn the contribution weights of different EEG channels. In terms of temporal information, LSTM is integrated into the model to identify the short-term coherence and long-term consistency features. To assess the effectiveness of LSSENet, a comprehensive set of experiments was carried on the AMIGOS dataset. The results show that LSSENet outperforms other EEG-based personality recognition methods, achieving the best classification average accuracy in every dimension of Big Five personality traits.