3D scatter plots are a powerful visualisation method by being able to represent 3 dimensions spatially. It can also enable the representation of additional dimensions, such as by using a colour map. An important issue with the current state of plotting software is the limited use of physical properties from the real world such as shadows to improve the effectiveness of the plots. A popular example is with the use of isometric axes in combination with same-sized points, which is equivalent to removing one whole dimension (depth perception). In static snapshot images, as found in digital and hard prints, as well with discrete data, additional cues such as movement are not present to mitigate for the loss of spatial information. In this paper we present a novel plotting framework that features a wide range of techniques to improve the information transfer from 3D scatterplots for multi-dimensional data. We evaluate the resulting plots by surveying 57 participants from an academic institution to get important insights on what makes 3D scatterplots effective in communicating data of more than two dimensions.
翻译:暂无翻译