Visualization of Diversity in Large Multivariate Data Sets

Abstract

(a) Very low diversity (b) Medium diversity (c) Very high diversity Fig. 1. Synthetic data sets of (a) very low-, (b) medium-, and (c) very high-diversity visualized using the Diversity Map representation. Each visualized data set contains 1000 objects and 6 attributes (columns from left to right: SAT verbal, SAT math, SAT writing, ethnicity, gender, and income level). The very high-diversity data set is 6 times more diverse than the very low-diversity one.Abstract-Understanding the diversity of a set of multivariate objects is an important problem in many domains, including ecology, college admissions, investing, machine learning, and others. However, to date, very little work has been done to help users achieve this kind of understanding. Visual representation is especially appealing for this task because it offers the potential to allow users to efficiently observe the objects of interest in a direct and holistic way. Thus, in this paper, we attempt to formalize the problem of visualizing the diversity of a large (more than 1000 objects), multivariate (more than 5 attributes) data set as one worth deeper investigation by the information visualization community. In doing so, we contribute a precise definition of diversity, a set of requirements for diversity visualizations based on this definition, and a formal user study design intended to evaluate the capacity of a visual representation for communicating diversity information. Our primary contribution, however, is a visual representation, called the Diversity Map, for visualizing diversity. An evaluation of the Diversity Map using our study design shows that users can judge elements of diversity consistently and as or more accurately than when using the only other representation specifically designed to visualize diversity.

Evolutionary Visual Exploration: Evaluation of an IEC Framework for Guided Visual Search

“…Black cells represent non-diverse generation bins where individuals have identical values for that diversity factor. This visualization is similar to the diversity map by Pham et al (2010) where rows are attributes of a multidimensional dataset and columns are attribute value buckets.…”

Toward understanding online sentiment expression: an interdisciplinary approach with subgroup comparison and visualization

“…Compared with the ''node-link diagram'' approach, it is more easily scrollable and adjustable. Typical examples in the ''tabular'' approach include PowerSetViewer Munzner et al (2005), TableLens Rao and Card (1995), Diversity Map Pham et al (2010) andLineUp Gratzl et al (2013). Pow-erSetViewer first indexes each incoming itemset with a horizontally one-dimensional array of cells, which, from left to right, increases the itemset order.…”

Rationale Visualization for Safety and Security

“…Aggregated views of data, like in (stacked) histograms and heat maps, are also unwanted, since both individual tuples and attributes are of interest. Diversity maps [PHJ*10] do visualize the diversity in a set of tuples, but not between tuples since they also use aggregation. Multidimensional icons like, Chernoff faces [Che73] and Star glyphs [War13], explicitly map attribute values, which results in an unnecessary cognitive and perceptual load when checking only for conflict and agreement.…”