在线学习环境中基于眼动特征情感识别研究

在e-Learning环境中,学习普遍存在“情感缺失”问题,该问题会导致学习效果和学习体验下降。在学习过程中及时识别学习者的情感状态是解决“情感缺失”的首要问题,情感识别技术在人机交互教学得到了广泛的应用,但仍然存在不少问题和挑战。随着眼动追踪技术的发展,将眼动信号引入情感识别成为研究的热点。针对当前国内外在e-Learning环境中基于眼动特征的相关研究进行综述,对相关研究中采用的眼动特征、机器学习方法以及涉及的学习过程进行分类、归纳及分析,归纳了五类学习过程研究中常用的眼动特征和识别算法。通过对应用在疲劳检测、健康医疗以及人机交互等相关领域中的眼动特征进行分析,对可借鉴至MOOC学习环境下情感识别的眼动特征进行汇总,并为下一步如何采用眼动特征在MOOC环境下进行情感识别研究提出建议。

Research on Emotion Recognition Based on Eye Movement Features ine-Learning Environment

In the e-Learning environment,there is a common problem of“affective deficiency”in learning,which will lead to a decline in learning effectiveness and learning experience.Therefore,identifying the learner’s emotional state in time in the learning process is the first issue to solve the“affective deficiency”.Emotion recognition technology is widely used in human-computer interactive teaching,but there are still many problems and challenges.With the development of eye tracking technology,the introduction of eye movement signals into emotion recognition has become a research hotspot.Various researches based on eye movement characteristics in e-Learning environment at home and abroad are summarized,and the characteristics of eye movement,the machine learning methods and the learning process are classified,summarized and analyzed.The eye movement features and recognition algorithms commonly used in the research are summarized on the basis of five types of learning process.By analyzing the eye movement features used in fatigue detection,health care,human-computer interaction and other related fields,the eye movement features that can be used for emotion recognition in the MOOC learning environment are summarized.It also puts forward suggestions on how to use the eye movement feature to conduct emotion recognition research in the MOOC environment.