Executive Summary CH2
Edited by Ben. Turn to ContentPage
Get the knowledge flowing and circulating! :)
目录
The science of analytical reasoning provides the reasoning framework upon which one can build both strategic and tactical visual analytics technologies for threat analysis, prevention, and response. This reasoning process is central to the analyst's task of applying human judgments to reach conclusions from a combination of evidence and assumptions. Analysis may require collaborative effort, especially in emergency response and border security contexts.
The goal of visual analytics is to facilitate this analytical reasoning process through the creation of software that maximizes human capacity to perceive, understand, and reason about complex and dynamic data and situations. It must build upon an understanding of the reasoning process, as well as an understanding of underlying cognitive and perceptual principles, to provide mission-appropriate interactions that allow analysts to have a true discourse with their information. The goal is to facilitate high-quality human judgment with a limited investment of the analysts' time.
Several actions are necessary to advance the science of analytical reasoning in support of visual analytics.
Recommendation
Build upon theoretical foundations of reasoning, sense-making, cognition, and perception to create visually enabled tools to support collaborative analytic reasoning about complex and dynamic problems.
To truly support the analytical reasoning process, we must enable the analyst to focus on what is truly important. We must support the processes involved in making sense of information and developing and evaluating alternative explanations. Tools and techniques must support both convergent thinking, which involves assembling evidence to find an answer, and divergent thinking, which involves thinking creatively to ensure that plausible alternatives have not been overlooked. These tools and techniques also must allow analysts to look at their problem at multiple levels of abstraction and support reasoning about situations that change over time, sometimes very rapidly. They must support collaboration and team-work, often among people with very different backgrounds and levels of expertise. Accomplishing this will require the development of theory to describe how interactive visual discourse works, both perceptually and cognitively, in support of analytical reasoning.
Recommendation
Conduct research to address the challenges and seize the opportunities posed by the scale of the analytic problem. The issues of scale are manifested in many ways, including the complexity and urgency of the analytical task, the massive volume of diverse and dynamic data involved in the analysis, and challenges of collaborating among groups of people involved in analysis, prevention, and response efforts.
The sheer volume and scale of data involved in the analytical process offer as many opportunities as they do challenges for visual analytics. A science of scalable, visually based analytical reasoning, or visual analytic discourse, must take the issue of scale into consideration. Different types of analytic discourse will be appropriate to different analytical tasks, based on the level of complexity of the task, the speed with which a conclusion must be reached, the data volumes and types, and the level of collaboration involved.
分析推理科学提供了一个推理框架，在此基础上可以构建用于威胁分析、预防和应对的战略和战术（tactical）视觉分析技术。这个推理过程是分析师的任务核心，即应用人类判断，从证据和假设的组合中得出结论。分析可能需要合作，特别是在应急响应和边境安全方面。
视觉分析的目标是通过创建软件来促进这种分析推理过程，最大限度地提高人类感知、理解和推理复杂动态数据和情况的能力。它必须建立在对推理过程的理解以及对潜在（underlying）认知和感知原则的理解之上，以提供适合任务的互动，使分析师可以对他们的信息进行真实的对话（discourse）。其目标是在有限的分析师的时间投入下，促进高质量的人类判断。
为了推进支持视觉分析的分析推理科学，需要采取一些行动。
建议
建立在推理、感知、认知和感知的理论基础上，创建可视化工具，支持对复杂和动态问题的协作分析推理。
为了真正支持分析推理过程，我们必须使分析师专注于真正重要的事情。我们必须支持理解信息、制定和评估替代解释所涉及的过程。工具和技术必须支持趋同（convergent）思维和发散（divergent）思维，前者涉及收集（assembling）证据以找到答案，后者涉及创造性思维以确保合理的（plausible）替代方案不会被忽视。这些工具和技术还必须允许分析人员在多个抽象层次上看待他们的问题，并支持对随时间变化的情况进行推理，有时变化非常快。他们必须支持协作和团队合作，通常是在具有不同背景和专业水平的人之间。要做到这一点，就需要发展理论来描述交互式视觉话语（discourse）在感知和认知方面是如何工作的，以支持分析推理。
建议
进行研究以应对分析问题规模带来的挑战并抓住（seize）机遇。规模问题表现在许多方面，包括分析任务的复杂性和紧迫性，分析中涉及的大量多样和动态数据，以及参与分析、预防和应对工作的人群之间合作的挑战。
分析过程中涉及的绝对（sheer）数据量和规模为视觉分析提供了与挑战一样多的机会。一门可扩展的、基于视觉的分析推理或视觉分析话语的科学，必须考虑到尺度问题。根据任务的复杂程度、得出结论的速度、数据量和类型以及所涉及的协作水平，不同类型的分析话语将适用于不同的分析任务。