返回
类型 基础研究 预答辩日期 2018-01-20
开始(开题)日期 2015-12-22 论文结束日期 2017-11-19
地点 机械楼南高厅 论文选题来源 国家自然科学基金项目     论文字数 12 (万字)
题目 基于用户认知的大数据可视化视觉呈现方法研究
主题词 大数据可视化,视觉呈现,图元关系,信息编码,交互设计
摘要 大数据时代带给我们的既是机遇也是挑战。高通道的科学实验、高速的科学计算、高分辨率的传感器以及错综复杂的网络环境共同促生了大数据时代的到来。有限的可视化展示页面和高维多元结构复杂的大数据之间的矛盾日益凸显,这个矛盾所造成的空间局限性是大数据可视化面临的根本问题。人位于可视化信息结构的顶端,是对大数据可视化中信息进行态势判断和行为决策的认知主体。因此,满足人的认知特点、采用多页面分层展示、最终实现信息的按需呈现是大数据可视化所追求的根本目标。 本研究以该目标为导向、以大数据信息特征和人的认知特征为切入点、以可视化页面的视觉呈现为研究对象、以人的生理和行为评价作为检验方法,系统地研究了基于认知的大数据可视化的视觉呈现方法。研究将理论研究、案例分析与实验观察和实例验证相结合,本研究的成果可以为大数据可视化提供方法指导。本文的主要研究内容和创新点包括: (1)以大数据可视化中人—信息交互多水平结构模型为研究架构,依次对大数据可视化的信息空间、精神空间、表征空间和交互空间进行深入探讨,以认知为导向来研究大数据可视化中的问题,并提出大数据可视化人机协同作业的复杂认知模型作为本研究的理论基础。 (2)从单个页面上的视觉维度映射和多个页面间的交互设计两个部分切入,以信息维度和视觉呈现的映射规律为导向来探讨相应的大数据可视化的视觉呈现设计。在页面呈现的宏观视角上对信息图元关系进行分类描述,探讨图元关系和信息维度之间的关系;在页面呈现的微观视角上阐述了定序和分类的维度表征及双维度的整合-分离规律。信息和视觉呈现之间的映射关系研究成果可以作为大数据可视化页面表征的设计方法。 (3)将人和信息之间沟通的交互设计作为大数据可视化的研究重点,以实现多页面间的信息有效融合和知识连贯性为目标,提出适用于大数据可视化的交互设计原则、交互设计维度和有效的交互设计策略,为大数据可视化的交互设计提供方法指导。 (4)首次将视觉动量概念引入到大数据可视化评价中,提出用以眼动追踪数据为基础的复合虚拟指标——视觉动量的评价方法,该方法可用来来量化评价多页面动态交互式可视化界面,并在行为实验中得到验证。该眼动评价方法具有一般工作环境下推广的可行性,可以作为大数据可视化设计的客观评价方法。
英文题目 RESEARCH ON THE VISUAL PRESENTATION METHOD OF BIG DATA VISUALIZATION BASED ON USER COGNITION
英文主题词 big data visualization, visual presentation, primitive relationship, information encoding, interactive design
英文摘要 Oppportunities and challenges are generated in the age of big data. The big data era was created by high-channel scientific experiments, high-speed scientific computing, high- resolution sensors and intricate network environment. The contradiction between the limited visual display space and high-dimensional and multivariant complex big data has become increasingly prominent. The problem of spatial limitation caused by this contradiction is the fundamental issue of big data visualization. Human being who is the cognitive subject to make decision of information situation stands at the top of the structure lf visual information. Therefore, the pursuit of big data visualization should be to satisfy people’s cognitive characteristics, and ultimately achieve the information presenting on demand by hierarchical displaying in multi pages. The study aims to solve the problem of spatial limitation and starts from the information characteristics of big data and the cognitive characteristics of human being. Visual presentation of visualization interface was taken as the research object and the human physiological and behavioral evaluation was taken as the test method, this study systematically discussed on the visual presentation methods of big data visualization based on cognition. The study combined theoretical research, cases study and experimental observations and design verification together. The results of the research can provide methodological guidance for big data visualization. The main research contents and innovations of this paper were listed in the following. (1)The human-information interactive multi-level structure model was taken as the research structure. This study discussed on the information space, mental space, repretention space and interaction space of big data visualization in order. Based on human’s cognition, this study put forward the complex cognitive model of human-computer collaboration of big data visualization. And it could be the theoretical basis of big data visualization research. (2) The visual presentation design was descibed in two parts, the visual dimension mapping on the single page and the interactive design between multi pages. It focused on the mapping rules of information dimensions and visual presentations. This study presented the primitive relationship in classification and the relationship between primitive and information dimensions in the macroscopic view,and the qualitative and quantitative dimension presentation and the integration-separation law between two dimensions in the microscopic view. The mapping relationship between information and visual presentation provided the design methods for big data visualization presentation. (3)Considering the interactive design between human and information as the research focus, and achieving the effective integration of information and knowledge consistency among multiple pages as the goal, the interactive design principles, interactive design dimensions and effective interactive design strategy of big data visualization were put forward in this study. It could provide the methodological guidance for big data visualization interactive design. (4)In the study of evaluation of big data visualization, a composite virtual indicator —visual momentum index based on eye tracking data was proposed to quantified evaluate the multi-page dynamic interactive visualization interface. And it was verified in behavior experiments. This physiological evaluation method has the feasibility to generalizate into general working environment and could be employed as an objective evaluation method of big data visualization design.
学术讨论
主办单位时间地点报告人报告主题
机械工程学院 2014.05.15 机械楼302 周蕾 界面元素布局设计的美度评价方法
机械工程学院 2014.09.12 机械楼南高厅 马青 图书馆资源与服务导览
机械工程学院 2014.11.07 机械楼302 李晶 均衡认知负荷的人机界面信息编码
机械工程学院 2014.09.25 机械楼302 章寅 基于固态纳米通道的单分子传感器设计与制造
机械工程学院 2015.06.29 机械楼246 周小舟 传统数据可视化与大数据可视化的比较
机械工程学院 2015.09.30 机械楼246 周小舟 大数据可视化
机械工程学院 2015.10.14 机械楼246 周小舟 可视化中的二维与三维显示
机械工程学院 2017.10.10 机械楼南高厅 周小舟 面向大数据的信息可视化设计方法研究
     
学术会议
会议名称时间地点本人报告本人报告题目
2014设计教育现场交流与专家论坛 2014.06.08 东南大学九龙湖校区圆形报告厅 科学研究过程——产品设计中的方法论
第六届中国信息融合大会 2014.11.07 南京国际会议大酒店 面向大数据的信息可视化研究
International Conference on Human-Computer Interaction 2017.07.14 Vancouver Convention Centre Smooth” or “Intermittent”? The Necessity of Halt in the Dynamic Visualization due to the Features of Working Memory
     
代表作
论文名称
Spatial Conformity Research of Temporal Order Information Presentation in Visualization Design
产品族设计DNA可遗传因子提取
“Smooth” or “Intermittent”? The necessity of halt in the dynamic visualization due to the features o
An Evaluation Method of Visualization Using Visual Momentum Based on Eye-Tracking Data
 
答辩委员会组成信息
姓名职称导师类别工作单位是否主席备注
韦志辉 正高 教授 博导 南京理工大学
姚志远 正高 教授 博导 南京航空航天大学
吴智慧 正高 教授 博导 南京林业大学
王兴松 正高 教授 博导 东南大学
陈南 正高 教授 博导 东南大学
      
答辩秘书信息
姓名职称工作单位备注
陈斌 其他 讲师 东南大学