| 钟建,唐明清,唐秋阳,李文博,许娜.汽车人机交互产品用户满意度度量方法研究[J].包装工程,2019,40(6):239-244. |
| 汽车人机交互产品用户满意度度量方法研究 |
| Method for Measuring User Satisfaction with Human-computer Interaction Products of Automobiles |
| 投稿时间：2018-11-12 修订日期：2019-03-20 |
| DOI：10.19554/j.cnki.1001-3563.2019.06.041 |
| 中文关键词: 面部表情识别 人机交互 可用性 满意度 度量模型 BP神经网络 |
| 英文关键词: facial expression recognition human-computer interaction availability satisfaction metric model, BP neural network |
| 基金项目: |
|
| 摘要点击次数: |
| 全文下载次数: |
| 中文摘要: |
| 目的 研究基于面部表情识别技术的用户满意度客观度量方法。方法 以两款车载信息系统为载体，以面部表情识别与BP神经网络算法为技术手段，设计用户分别与两款系统进行人机交互的实验，建立用户面部表情与用户主观满意度的映射模型，并进行用户满意度预测，对比模型预测值与用户主观量表值，分析得出模型的预测能力，验证度量方法的可行性。结论 该模型对用户满意度的预测值与用户主观满意度值的整体均方误差为0.165，实现了在较小误差范围内的准确预测。模型通过识别用户与产品进行人机交互时的面部表情，能有效客观地度量用户对产品的满意度。 |
| 英文摘要: |
| The paper aims to study objective measurement method of user satisfaction based on facial expression recognition technology. Two vehicle-mounted information systems were taken as the carriers to design experiment for human-computer interaction with the two systems respectively through facial expression recognition and BP neural network algorithm. The user's facial data of the first system was used to establish the mapping model between user’s facial expression and user's subjective satisfaction, and the model was used to predict the user's satisfaction of the second system. The model predictive value was compared with the user subjective scale value, and the predictive ability of the model was analyzed to verify feasibility of the above measures. The overall mean square error of the model for predicting user satisfaction and user subjective satisfaction is 0.165, which realizes the accurate prediction within the smaller error range. By identifying the facial expression information of users when interacting with the product, the model could effectively and objectively measure users' satisfaction with the product. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
