IEEE Trans Vis Comput Graph. 2018 Jan;24(1):340-350. doi: 10.1109/TVCG.2017.2745279. Epub 2017 Aug 29.
During asynchronous collaborative analysis, handoff of partial findings is challenging because externalizations produced by analysts may not adequately communicate their investigative process. To address this challenge, we developed techniques to automatically capture and help encode tacit aspects of the investigative process based on an analyst's interactions, and streamline explicit authoring of handoff annotations. We designed our techniques to mediate awareness of analysis coverage, support explicit communication of progress and uncertainty with annotation, and implicit communication through playback of investigation histories. To evaluate our techniques, we developed an interactive visual analysis system, KTGraph, that supports an asynchronous investigative document analysis task. We conducted a two-phase user study to characterize a set of handoff strategies and to compare investigative performance with and without our techniques. The results suggest that our techniques promote the use of more effective handoff strategies, help increase an awareness of prior investigative process and insights, as well as improve final investigative outcomes.
在异步协作分析中，部分发现的交接具有挑战性，因为分析人员生成的外部表示形式可能无法充分传达他们的调查过程。为了解决这个挑战，我们开发了技术，以便根据分析师的交互自动捕获和帮助编码调查过程的隐性方面，并简化交接批注的显式创作。我们设计这些技术来调节对分析覆盖范围的意识，支持通过批注显式交流进度和不确定性，并通过回放调查历史进行隐式交流。为了评估我们的技术，我们开发了一个交互式可视化分析系统 KTGraph，它支持异步调查文档分析任务。我们进行了两阶段的用户研究，以描述一组交接策略，并比较有无我们技术的调查性能。结果表明，我们的技术促进了更有效的交接策略的使用，有助于提高对先前调查过程和见解的认识，并改善最终的调查结果。
