Perin Charles, Boy Jeremy, Vernier Frederic
IEEE Comput Graph Appl. 2016 Sep-Oct;36(5):38-49. doi: 10.1109/MCG.2016.100.
To address the limitations of traditional line chart approaches, in particular rank charts (RCs) and score charts (SCs), a novel class of line charts called gap charts (GCs) show entries that are ranked over time according to a performance metric. The main advantages of GCs are that entries never overlap (only changes in rank generate limited overlap between time steps) and gaps between entries show the magnitude of their score difference. The authors evaluate the effectiveness of GCs for performing different types of tasks and find that they outperform standard time-dependent ranking visualizations for tasks that involve identifying and understanding evolutions in both ranks and scores. They also show that GCs are a generic and scalable class of line charts by applying them to a variety of different datasets.
为解决传统折线图方法的局限性，尤其是排名图（RC）和分数图（SC）的局限性，一类名为差距图（GC）的新型折线图展示了根据绩效指标随时间排序的条目。差距图的主要优点是条目从不重叠（只有排名变化会在时间步之间产生有限的重叠），并且条目之间的差距显示了它们得分差异的大小。作者评估了差距图在执行不同类型任务时的有效性，发现对于涉及识别和理解排名及分数演变的任务，它们优于标准的时间相关排名可视化方法。他们还通过将差距图应用于各种不同的数据集，表明差距图是一类通用且可扩展的折线图。
