Browsing by Author "Aupetit, Michaël"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    ClustMe: A Visual Quality Measure for Ranking Monochrome Scatterplots based on Cluster Patterns
    (The Eurographics Association and John Wiley & Sons Ltd., 2019) Abbas, Mostafa M.; Aupetit, Michaël; Sedlmair, Michael; Bensmail, Halima; Gleicher, Michael and Viola, Ivan and Leitte, Heike
    We propose ClustMe, a new visual quality measure to rank monochrome scatterplots based on cluster patterns. ClustMe is based on data collected from a human-subjects study, in which 34 participants judged synthetically generated cluster patterns in 1000 scatterplots. We generated these patterns by carefully varying the free parameters of a simple Gaussian Mixture Model with two components. and asked the participants to count the number of clusters they could see (1 or more than 1). Based on the results, we form ClustMe by selecting the model that best predicts these human judgments among 7 different state-of-the-art merging techniques (DEMP). To quantitatively evaluate ClustMe, we conducted a second study, in which 31 human subjects ranked 435 pairs of scatterplots of real and synthetic data in terms of cluster patterns complexity. We use this data to compare ClustMe's performance to 4 other state-of-the-art clustering measures, including the well-known Clumpiness scagnostics. We found that of all measures, ClustMe is in strongest agreement with the human rankings.
  • Thumbnail Image
    Item
    A Force-Directed Power Diagram Approach for Interactive Voronoi Treemaps
    (The Eurographics Association, 2020) Abuthawabeh, Ala; Aupetit, Michaël; Kerren, Andreas and Garth, Christoph and Marai, G. Elisabeta
    Voronoi treemaps represent weighted hierarchical data as nested Voronoi diagram partitions with cells' area proportional to the weights. Current techniques to compute them propose static visualizations which can be used for reporting, or dynamic one to capture data update. However, no ideal solution exists yet to interactively rearrange the treemap layout, for instance for a data journalist to tell a story, or for a scientist to create data categorization. We propose a new way to get an interactive Voronoi treemap, where a child cell can be moved by drag-and-drop within a parent cell attempting to preserve both stability (position) and weight (area) during the move. We use a force-directed approach applied to the dual circles of the Power cells to guide the computation of the Power diagram under the hood. Our preliminary quantitative experiments show the force-directed approach provides areas with 10% weighted average error, which is an order of magnitude higher than standard static approaches, but qualitative observations show that it gives a more predictable and smoother interaction, and a direct control over the stability of the remaining cells. Assuming the user would focus less on getting high accuracy of the areas than keeping a good and stable overview of the treemap while dragging a cell, the force-directed approach appears to be a valuable option to explore further. We also discovered a trade-off between stability and accuracy and the force-directed approach lets the user control it directly.
  • Thumbnail Image
    Item
    The Human User in Progressive Visual Analytics
    (The Eurographics Association, 2019) Micallef, Luana; Schulz, Hans-Jörg; Angelini, Marco; Aupetit, Michaël; Chang, Remco; Kohlhammer, Jörn; Perer, Adam; Santucci, Giuseppe; Johansson, Jimmy and Sadlo, Filip and Marai, G. Elisabeta
    The amount of generated and analyzed data is ever increasing, and processing such large data sets can take too long in situations where time-to-decision or fluid data exploration are critical. Progressive visual analytics (PVA) has recently emerged as a potential solution that allows users to analyze intermediary results during the computation without waiting for the computation to complete. However, there has been limited consideration on how these techniques impact the user. Based on discussions from a Dagstuhl seminar held in October 2018, this paper characterizes PVA users by their common roles, their main tasks, and their distinct focus of analysis. It further discusses cognitive biases that play a particular role in PVA. This work will help PVA visualization designers in devising systems that are tailored for their specific target users and their characteristics.