Object Reference in Task-Oriented Keyboard Dialogues
Object Reference in Task-Oriented Keyboard Dialogues
复制标题
面向任务的键盘对话中的对象引用
DOI：
10.1007/bfb0052324
复制
发表时间：
1995
期刊：
影响因子：
--
通讯作者：
A. Cremers
中科院分区：
文献类型：
--
作者：
A. Cremers
在DenK项目中，开发了一个多模态界面，其中自然语言与图形交互相结合。对于这个界面的设计，通过自然语言和手势来收集关于人类如何在面向任务的环境中引用对象的知识。在本文中，我们报告了一个关于在面向任务的键盘对话中引用行为的实验结果。结果与我们之前进行的口语对话实验的结果进行了比较。这些差异都被发现与所谓的最小合作总努力原则有关，也就是说，在可用模式的限制下，参与者的目标是花费尽可能少的总努力在另一方面提到某个物体，另一方面识别物体。在此基础上，提出了包括类型化自然语言在内的多模态界面设计建议。
In the DenK project a multimodal interface is developed where natural language is combined with graphical interaction. For the design of this interface, knowledge is collected about how humans refer to objects in a task-oriented environment, by means of natural language and gestures. In this paper we report results of an experiment concerning referring behaviour in tasd-oriented keyboard dialogues. The results are compared with those of an earlier experiment we performed with spoken dialogues. The differences were all found to be related to the so-calledprinciple of minimal cooperative total effort, which says that, within the limitations of the available modalities, the participants aim at spending as little total effort as possible on referring to a certain object on the other hand, and on identifying the object on the other hand. Based on the results, we formulate recommendations for the design of multimodal interfaces which include typed natural language.
虚拟演员的头眼协同运动模型
DOI：
--
发表时间：
2005
期刊：
Trans. IEICE Japan Vol.J88-D-II, No.3
影响因子：
--
作者：
Soh Masuko;Junichi Hoshino
通讯作者：
Junichi Hoshino