利用实验室现有设备采集测试者在4种情绪(Joy、Anger、Sadness、Pleasure)下的3种生理信号(心电、皮电及呼吸)。在此基础上,基于相关算法对采集的生理信号提取混沌特征值。以混沌特征作为模式识别的特征参数,采用C5.0决策树进行情绪识别。研究结果表明在情绪识别方面,基于3种生理信号比1种生理信号具有更高的识别率。
Three kinds of physiological signals ( ECG, SC and RSP ) of one subject are collected under four kinds of emotions such as joy, anger, sadness and pleasure by using an existing equipment in the laboratory .On this basis, chaotic characteristic values are ex-tracted from the collected physiological signals based on the relevant algorithms .C5.0 decision tree is used to make emotion recognition by taking chaotic characteristics as the characteristic parameters of pattern recognition .The research results show the recognition rate with three kinds of physiological signals is higher more than that with one kind of physiological signal in emotional recognition .