Comparative Evaluation of Animated Scatter Plot Transitions

Nils Rodrigues,Frederik L. Dennig,Vincent Brandt,Daniel A. Keim,Daniel Weiskopf

2024-01-10

Abstract:Scatter plots are popular for displaying 2D data, but in practice, many data sets have more than two dimensions. For the analysis of such multivariate data, it is often necessary to switch between scatter plots of different dimension pairs, e.g., in a scatter plot matrix (SPLOM). Alternative approaches include a "grand tour" for an overview of the entire data set or creating artificial axes from dimensionality reduction (DR). A cross-cutting concern in all techniques is the ability of viewers to find correspondence between data points in different views. Previous work proposed animations to preserve the mental map between view changes and to trace points as well as clusters between scatter plots of the same underlying data set. In this paper, we evaluate a variety of spline- and rotation-based view transitions in a crowdsourced user study focusing on ecological validity. Using the study results, we assess each animation's suitability for tracing points and clusters across view changes. We evaluate whether the order of horizontal and vertical rotation is relevant for task accuracy. The results show that rotations with an orthographic camera or staged expansion of a depth axis significantly outperform all other animation techniques for the traceability of individual points. Further, we provide a ranking of the animated transition techniques for traceability of individual points. However, we could not find any significant differences for the traceability of clusters. Furthermore, we identified differences by animation direction that could guide further studies to determine potential confounds for these differences. We publish the study data for reuse and provide the animation framework as a D3.js plug-in.

Human-Computer Interaction

What problem does this paper attempt to address?

### The Problem the Paper Attempts to Solve

This paper attempts to address the issue of maintaining the correspondence between data points when switching views between different scatter plots in multidimensional datasets, so that users can effectively track individual data points or clusters. Specifically:

1. **View Switching Problem in Multidimensional Data Analysis**: When analyzing multidimensional data, it is often necessary to switch between different scatter plots to observe the impact of different dimensions on the data. However, this switching can make it difficult for users to find the same data points or clusters in different views, thereby affecting the coherence of data understanding and analysis.

2. **Application of Animation Techniques**: To help users maintain the tracking of data points during view switching, the researchers propose using animation techniques for smooth transitions. These animation techniques include spline-based and rotation-based methods.

3. **Evaluating the Effectiveness of Animation Techniques**: The paper evaluates the effectiveness of various animation techniques in tracking individual data points and clusters through a crowdsourced user study. The study aims to determine which animation techniques are most helpful in maintaining the correspondence of data points between different scatter plots, thereby improving the efficiency and accuracy of data analysis.

### Research Background

- **Limitations of Scatter Plots**: Although scatter plots are effective tools for displaying 2-dimensional data, handling multidimensional data requires switching between multiple scatter plots, which increases the cognitive load on users.
- **Existing Solutions**: Some existing methods, such as "grand tour" and dimensionality reduction techniques, can help users get an overview of the entire dataset, but these methods face challenges in interpretability and tracking original data attributes.
- **Advantages of Animation Techniques**: Animation techniques can preserve users' "mental map," helping them track data points and clusters during view switching, thereby improving the coherence and accuracy of data analysis.

### Research Objectives

- **Evaluating Animation Techniques**: Through user studies, evaluate the effectiveness of different types of animation techniques in tracking individual data points and clusters.
- **Improving Data Analysis Efficiency**: Identify the most effective animation techniques to help users more efficiently track data points and clusters in multidimensional data analysis.
- **Ecological Validity**: Ensure that the research results have ecological validity in practical applications, not just theoretical validation in a laboratory environment.

### Main Contributions

- **User Study**: Designed and implemented a pre-registered crowdsourced user study involving 170 participants, quantitatively evaluating the effectiveness of different animation techniques.
- **Animation Framework**: Developed a D3.js-based animation framework plugin and provided research data and interactive demonstrations for other researchers and developers to use.

Through this research, the authors hope to provide practical guidance for multidimensional data visualization and analysis, helping users more effectively understand and analyze complex datasets.