Deconstructing Categorization in Visualization Recommendation: A Taxonomy and Comparative Study

TL;DR

This article explores the efficacy of recommendation categories by formalizing a taxonomy of common categories and developing a system, Frontier, that implements these categories and evaluates workflow strategies adopted by users and how categories influence those strategies.

Abstract

Visualization recommendation (VisRec) systems provide users with suggestions for potentially interesting and useful next steps during exploratory data analysis. These recommendations are typically organized into categories based on their analytical actions, i.e., operations employed to transition from the current exploration state to a recommended visualization. However, despite the emergence of a plethora of VisRec systems in recent work, the utility of the categories employed by these systems in analytical workflows has not been systematically investigated. Our paper explores the efficacy of recommendation categories by formalizing a taxonomy of common categories and developing a system, Frontier, that implements these categories. Using Frontier, we evaluate workflow strategies adopted by users and how categories influence those strategies. Participants found recommendations that add attributes to enhance the current visualization and recommendations that filter to sub-populations to be comparatively most useful during data exploration. Our findings pave the way for next-generation VisRec systems that are adaptive and personalized via carefully chosen, effective recommendation categories.

Figures (7)

• Figure 1: A screenshot of Frontier with a dataset containing college information. Starting from the Current View displaying a scatterplot of AverageCost versus SATAverage on the left, the user finds an interesting visualization recommended through the Enhance category highlighting the two distinct clusters for Private and Public FundingModels (shown with a red border). This recommendation is generated from the Current View, further "enhanced" by adding FundingModel to the color channel.
• Figure 2: A taxonomy of common analytical actions used in recommending visualizations for visual analysis. The analytical actions are indicated in blue.
• Figure 3: Operational actions represent transitions through the attribute and value hierarchies.
• Figure 4: Frontier consists of four areas: Control Panel (A), Current View (B), Category Menu (C), and Recommendations Panel (D).
• Figure 5: Examples of various recommendation categories implemented in Frontier. (A) Correlation generates scatterplots with bivariate relationships between quantitative fields ranging from high to low correlation. (B) Distribution shows the possible univariate distributions from the dataset ranging from skewed to normal distributions. In the following examples, the current view is shown on the left, with the corresponding recommendations shown on the right. (C) Generalize shows possible visualizations when one attribute or filter from the current view is removed (removed attributes shown with strikethroughs). (D) Similarity highlights data patterns ranging from most to least similar to the current view. (E) Pivot shows possible visualizations that can be constructed if one of the current attributes is changed to another (changed attributes shown in blue). (F) Enhance shows possible visualizations when an additional attribute is added to the current view (additional attributes shown in blue). (G) Filter displays the data subsets that can be constructed from the current view when a filter is applied.