Task-oriented data classification and design of choropleth maps (aChor)

Choropleth地图可能是最常用的主题地图类型。为了获得更好的概览和更快的可读性，通常会提前对显示的属性值进行分类。软件包中常用和实现的分类方法（如equidistance、Quantile、Jenks等）是数据驱动的，即仅根据原始值的当前频率分布来确定区间。在沿着数轴使用这种除法时，与许多应用程序相关的底层数据的空间上下文完全被忽略了。在这种情况下，关于空间关系或模式的信息（甚至是想要的地图“信息”）可能会丢失；在实践中（在最好的情况下），从几个分类选项中执行手动-主观选择。因此，本项目提倡以任务为导向的方法。虽然不可能预先为每个队列分配特定的任务或“消息”，但有足够的绝对数量的应用程序来证明以下开发是合理的。该项目将开发和测试方法，为可通过使用地形图执行的概要任务产生最佳数据分类。这些任务包括检测多边形、热点和冷点、全局或局部极值以及聚类区域之间的值差异。在创建类之后，这些类必须通过合适的配色方案进行可视化，以充分保留（甚至强调）相邻多边形或多边形组之间的原始属性值差异。将特别关注将值差转换为（比例）色差。迄今为止，考虑到具体使用目的的配色方案的选择和评价只在少数出版物中讨论过。最后，问题仍然是，面向任务的数据分类和相应的可视化的潜在附加价值是否真的被用户感知到，从而能够导致良好的决策。为此，进行了全面的经验用户研究，也考虑了数据分类步骤的各种参数。

Choropleth maps are the type of thematic maps that are probably most frequently used. To obtain a better overview and faster readability, the displayed attribute values are often classified in advance.The classification methods commonly used and implemented in software packages (such as equidistance, Quantile, Jenks, etc.) are data-driven, i.e., the intervals are determined solely on the basis of the present frequency distribution of the original values. The spatial context of the underlying data, which is relevant for many applications, is completely neglected when using such a division along the number line. With that information about spatial relationships or patterns (or even desired "messages" of a map) might be lost; in practice (and in a best case), a manual-subjective selection out of several classification options is performed.Therefore, in this project a task-oriented approach is advocated. Although it is not possible to assign a specific task or "message" to each choropleth in advance, there are enough applications in absolute numbers to justify the following developments.The project will develop and test methods that produce optimal data classifications for synoptic tasks which can be performed by using choropleth maps. These tasks include the detection of values differences between polygons, hot and cold spots, global or local extreme values as well as cluster regions.After the creation of classes those have to be visualized by a suitable color scheme, that adequately preserves (or even stresses) the original attribute value differences between adjacent polygons or groups of polygons. A special focus will be laid on the transformation of value differences into (proportional) color differences. So far, the selection and evaluation of a color scheme taking account of specific use purposes have been treated only in a few publications.Finally, the question remains whether the potential added value of a task-oriented data classification and a corresponding visualization is actually perceived by the user - and can thus lead to good decisions. For this purpose, comprehensive empirical user studies are carried out, that also take into account various parameters of the data classification step.