| 摘 要: 针对人机交互中用户认知模糊与表达不准确等问题,提出一种基于生理信号的PAD(Pleasure,Arousal,Dominance)多维情感预测方法对用户情感进行预测。首先,确定眼动信号指标和PAD情感量表,以门户网站为实验样本开展情感测量实验,被试者需按要求完成浏览任务,并记录眼动数据。其次,通过PAD情感量表获取被试者在网页界面中的多维情感值。最后,利用偏最小二乘回归法建立关系模型,探索眼动数据与PAD多维情感值之间的关系,并验证模型的有效性和适用性。研究结果表明,该情感预测模型的Sig.值均大于0.05,具有较高的预测能力,能准确预测用户对网页界面的情感偏好。 |
| 关键词: 生理信号;PAD情感;偏最小二乘回归;情感预测 |
|
中图分类号: TP39
文献标识码: A
|
|
| Research on PAD Multi-dimensional Emotion Prediction Method Based on Physiological Signals |
|
HE Jiale, ZHANG Jianmin
|
(School of Mechanical Engineering, Guizhou University, Guiyang 550000, China)
1138595904@qq.com; zminmindebox@126.com
|
| Abstract: Aiming at the issues such as user cognitive fuzziness and inaccurate expression in human-computer interaction, this paper proposes a PAD (Pleasure, Arousal, Dominance) multi-dimensional emotion prediction method based on physiological signals to predict user emotions. Firstly, eye-tracking signal indicators and PAD emotion scales are determined. Emotion measurement experiments are conducted using a portal website as the experimental sample, where participants are required to complete browsing tasks as instructed and eye-tracking data is recorded. Secondly, multi-dimensional emotional values of participants on the webpage interface are obtained through the PAD emotion scale. Lastly, a relationship model is established using Partial Least Squares Regression to explore the relationship between eye-tracking data and PAD multi-dimensional emotional values, and to validate the effectiveness and applicability of the model. The research results show that the Sig. values of this emotion prediction model are all greater than 0. 05, indicating good predictive capability to accurately predict users ' emotional preferences towards webpage interfaces. |
| Keywords: physiological signals; PAD emotions; Partial Least Squares Regression; emotion prediction |
