计算机与现代化 ›› 2013, Vol. 1 ›› Issue (2): 120-122.doi: 10.3969/j.issn.1006-2475.2013.02.029

• 人工智能 • 上一篇    下一篇

基于自适应邻域禁忌搜索的情感生理信号特征选择研究

莫建麟1,王玉晶2   

  1. 1.阿坝师范高等专科学校电子信息工程系,四川汶川623002;2.西南财经大学天府学院,四川绵阳621000
  • 收稿日期:2012-10-19 修回日期:1900-01-01 出版日期:2013-02-27 发布日期:2013-02-27

Research on Emotion Physiological Signals Feature Selection Based on Adaptive Neighborhood TS

MO Jian-lin1, WANG Yu-jing2   

  1. 1. Department of Electronic Information Engineering, Aba Teachers College, Wenchuan 623002, China;2. Tianfu College of SWUFE, Mianyang 621000, China
  • Received:2012-10-19 Revised:1900-01-01 Online:2013-02-27 Published:2013-02-27

摘要: 在生理信号的情感识别中,有效特征子集的选择是识别情感的重要步骤,为解决这一问题,提出一种改进的邻域构造方法。通过自适应地产生邻域元素,较好地解决了情感生理信号的特征选择问题。仿真实验结果表明,该方法找到了有效的特征子集,达到了很好的情感识别效果,是可行和有效的。

关键词: 情感识别, 禁忌搜索, 邻域规则, 特征选择

Abstract: In emotion recognition of physiological signals, selection of valid feature subset is the important step to identify emotion. An improved neighborhood structure by adaptive neighborhood elements is proposed to solve the problem about feature selection of emotion physiology signal. Simulation shows that the method of finding valid feature subset is feasible and effective.