Orko: Facilitating Multimodal Interaction for Visual Exploration and Analysis of Networks
Orko: Facilitating Multimodal Interaction for Visual Exploration and Analysis of Networks
复制标题
DOI：
10.1109/tvcg.2017.2745219
复制
发表时间：
2018
影响因子：
5.2
通讯作者：
Arjun Srinivasan;J. Stasko
中科院分区：
文献类型：
--
作者：
Arjun Srinivasan;J. Stasko
数据可视化系统主要是为基于WIMP的直接操作界面开发的。直到最近才开始出现其他形式的交互，例如自然语言或基于触摸的交互，尽管通常只能独立操作。之前的评估自然语言界面的可视化显示了潜在的价值，结合直接操作和自然语言作为互补的交互技术。我们假设，真正的多模态界面的可视化，那些通过自然语言和基于触摸的直接操作输入，提供用户的表达自由，可以提供一个有效的和引人入胜的用户体验。不幸的是，然而，很少的工作已经做了探索这样的多模态可视化界面。为了解决这个问题，我们创建了一个架构和一个名为Orko的原型可视化系统，它可以促进自然语言和直接操作输入。具体来说，Orko专注于网络可视化领域，该领域在很大程度上依赖于基于WIMP的界面和直接操作交互，并且很少或根本没有探索自然语言交互的研究。我们报告Orko的初步评估研究的结果，并使用我们的观察来讨论未来在多模态网络可视化界面工作的机遇和挑战。
Data visualization systems have predominantly been developed for WIMP-based direct manipulation interfaces. Only recently have other forms of interaction begun to appear, such as natural language or touch-based interaction, though usually operating only independently. Prior evaluations of natural language interfaces for visualization have indicated potential value in combining direct manipulation and natural language as complementary interaction techniques. We hypothesize that truly multimodal interfaces for visualization, those providing users with freedom of expression via both natural language and touch-based direct manipulation input, may provide an effective and engaging user experience. Unfortunately, however, little work has been done in exploring such multimodal visualization interfaces. To address this gap, we have created an architecture and a prototype visualization system called Orko that facilitates both natural language and direct manipulation input. Specifically, Orko focuses on the domain of network visualization, one that has largely relied on WIMP-based interfaces and direct manipulation interaction, and has little or no prior research exploring natural language interaction. We report results from an initial evaluation study of Orko, and use our observations to discuss opportunities and challenges for future work in multimodal network visualization interfaces.