Analytical reasoning techniques are methods by which users explore their data to obtain insight and knowledge that can directly support situational awareness and decision making. Recently, the analytical reasoning process has been augmented through the use of interactive visual representations and tools which utilize cognitive, design and perceptual principles. These tools are commonly referred to as visual analytics tools, and the underlying methods and principles have roots in a variety of disciplines. This chapter provides an introduction to young researchers as an overview of common visual representations and statistical analysis methods utilized in a variety of visual analytics systems. The application and design of visualization and analytical algorithms are subject to design decisions, parameter choices, and many conflicting requirements. As such, this chapter attempts to provide an initial set of guidelines for the creation of the visual representation, including pitfalls and areas where the graphics can be enhanced through interactive exploration. Basic analytical methods are explored as a means of enhancing the visual analysis process, moving from visual analysis to visual analytics.
Table of Contents
Visual Representations and Analysis
About the Author(s)Ross Maciejewski
, Purdue University
Ross Maciejewski received his PhD in 2009 from Purdue University for his thesis "Exploring Multivariate Data through the Application of Visual Analytics." Currently, he is a visiting assistant professor at Purdue University working as a member of the visual analytics for command, control, and interoperability environments Department of Homeland Security Center of Excellence. His research interests include visual analytics, illustrative visualization, volume rendering, non-photorealistic rendering and geovisualization.
This book presents mathematical, statistical, and visualization techniques for analyzing and communicating multivariate data.
The book has two parts. Chapters 1 to 3 present foundations such as data types, color schemes to encode data, and data preconditioning methods to transform data into representations with statistical properties more suitable for subsequent analysis and visualization techniques. The second part (chapter 4) presents visualization and visual analysis techniques for multivariate data.
At around 75 pages, the book is very compact; however, it gives a good overview of the domain. It is well structured and many of the techniques are illustrated with interesting examples. In many instances, useful practical advice is given, for example, on how to choose a color scheme. I especially like that the author provides some insight about which techniques are used in which application scenarios...
The compact nature of the book limits the amount of algorithmic and implementation detail. Most of the mathematical and statistical techniques are correctly captured by formulas, but many undergraduate students new to this field might struggle to translate them into appropriate algorithms. Students with programming experience would benefit from more references relevant to an implementation. Students without programming experience would benefit from references to tools and libraries implementing the described algorithms. The summary at the end mentions a few tools (MATLAB, R, and Mondrian), but I would have liked to see this information earlier and in combination with a more hands-on approach.
The introduction of the relatively new power transform is interesting, and chapter 4 includes examples for using this transform to obtain statistical properties necessary for subsequent analytical analysis. The power transform skews the data, which impacts perception and cognition by the user, for example, when using color mapping. I would have liked to see more of a critical discussion in this and other instances, and some consideration of the potential pitfalls of using certain techniques. A good example for this is given at the bottom of page 48.
The pictures and illustrations are very helpful in understanding the text. Unfortunately, however, they are only in black and white. In many cases, color is essential for understanding the images, such as the color scales in chapter 2 and Figures 4.9 and 4.10. The labeling is sometimes very small and occasionally unreadable (see Figure 4.12). A simple solution would be to provide a link to a Web page with high-resolution color images. This could then also be used to provide videos demonstrating interactive visualizations and animations of time series data.
The introductory section would benefit from expansion. For example, chapter 2 talks about color scales, but not about visual attributes in general, which are important when designing effective visualizations. The techniques presented in chapter 4 make use of features such as colors, shapes, textures, and length on a scale.
There are a few typographical errors in the book. For example, the author states that the recommended range for alpha ranges from .05 to 0.3 alphas; this doesn’t make sense.
I believe this book is most suited for students who want to get a quick introduction to the domain, or as preparation for a research project. For students with little background in visualization or statistics, a more traditional textbook would provide a gentler introduction. Overall, I enjoyed reading the book and I found it helpful and easy to read. I hope future editions of this book will come with a CD or link to a Web page with suitable resources.Burkhard Wuensche - ACM Computing Reviews