Graph analysis is an effective approach to reveal the complex relationships in data set, and usually visualization plays a significant role during this process. The graph visualization techniques based on the simpli-fication technology can reduce the visual clutter as well as the complexity of the graph layout, enhance the user's cognition, and thus have great advantage while analyzing various kinds of complex relationships. According to the actual analytical needs in food safety fields, this paper, based on the traditional force directed graph layout, proposes a transformation-based graph called TransGraph, which integrates some simplification approaches in-cluding graph filtering, node clustering, fisheye distortion and some effective interactive exploration techniques like overview+detail, focus+context, etc. TransGraph consists of donuts, sunburst and other visual elements to make it possible for reflecting the data distribution and hierarchical structure of current focus while displaying the overall network relationship among nodes. Users can also view the details of the relations or the node similarity situation interactively according to the user interest. Finally, a visual analytic system called PestResiTGVis is provided based on the simulated data sets of pesticide residues in fruits and vegetables. The experimental result shows that TransGraph highlights the key regulatory objects and comprehensively demonstrate the related infor-mation, and thus can effectively assist the relevant authorities to make corresponding decisions.%图分析是揭示数据中复杂关联关系的一种有效手段,而可视化通常为该过程的核心组成部分.基于简化技术的图可视化方法可以降低图布局复杂度,减少视觉杂乱,提升用户体验,在分析复杂关联时有着重要的价值和优势.针对食品安全领域内的实际关联分析需求,在传统力导引图布局的基础上综合采用图过滤、节点聚类、鱼眼视图变换以及交互式分层探索等多种图简化技术,并结合donut圆环、放射环等可视化元素,提出了一种基于变换的可视分析关联图 TransGraph,支持在对比展示数据关联的同时展现当前关注节点的数据分布和层次结构,并根据用户关注度交互式地逐层展示图的细节及节点相似性.基于水果蔬菜中的农药残留模拟数据集,设计并实现了一个可视分析系统 PestResiTGVis,实验结果表明,采用 TransGraph能够突出重点监管对象、全面地展现关联信息,从而有效地辅助相关监管部门及分析人员制定决策.
展开▼
