2022-08-25T08:27:36Z, Cakmak, Eren, Fuchs, Johannes, Jäckle, Dominik, Schreck, Tobias, Brandes, Ulrik, Keim, Daniel A.

Many data analysis problems rely on dynamic networks, such as social or communication network analyses. Providing a scalable overview of long sequences of such dynamic networks remains challenging due to the underlying large-scale data containing elusive topological changes. We propose two complementary pixel-based visualizations, which reflect occurrences of selected sub-networks (motifs) and provide a time-scalable overview of dynamic networks: a network-level census (motif significance profiles) linked with a node-level sub-network metric (graphlet degree vectors) views to reveal structural changes, trends, states, and outliers. The network census captures significantly occurring motifs compared to their expected occurrences in random networks and exposes structural changes in a dynamic network. The sub-network metrics display the local topological neighborhood of a node in a single network belonging to the dynamic network. The linked pixel-based visualizations allow exploring motifs in different-sized networks to analyze the changing structures within and across dynamic networks, for instance, to visually analyze the shape and rate of changes in the network topology. We describe the identification of visual patterns, also considering different reordering strategies to emphasize visual patterns. We demonstrate the approach's usefulness by a use case analysis based on real-world large-scale dynamic networks, such as the evolving social networks of Reddit or Facebook.

2021-02, Cakmak, Eren, Schlegel, Udo, Jäckle, Dominik, Keim, Daniel A., Schreck, Tobias

The overview-driven visual analysis of large-scale dynamic graphs poses a major challenge. We propose Multiscale Snapshots, a visual analytics approach to analyze temporal summaries of dynamic graphs at multiple temporal scales. First, we recursively generate temporal summaries to abstract overlapping sequences of graphs into compact snapshots. Second, we apply graph embeddings to the snapshots to learn low-dimensional representations of each sequence of graphs to speed up specific analytical tasks (e.g., similarity search). Third, we visualize the evolving data from a coarse to fine-granular snapshots to semi-automatically analyze temporal states, trends, and outliers. The approach enables us to discover similar temporal summaries (e.g., reoccurring states), reduces the temporal data to speed up automatic analysis, and to explore both structural and temporal properties of a dynamic graph. We demonstrate the usefulness of our approach by a quantitative evaluation and the application to a real-world dataset.

2020, Cakmak, Eren, Jäckle, Dominik, Schreck, Tobias, Keim, Daniel A.

Presenting long sequences of dynamic graphs remains challenging due to the underlying large-scale and high-dimensional data. We propose dg2pix, a novel pixel-based visualization technique, to visually explore temporal and structural properties in long sequences of large-scale graphs. The approach consists of three main steps: (1) the multiscale modeling of the temporal dimension; (2) unsupervised graph embeddings to learn low-dimensional representations of the dynamic graph data; and (3) an interactive pixel-based visualization to simultaneously explore the evolving data at different temporal aggregation scales. dg2pix provides a scalable overview of a dynamic graph, supports the exploration of long sequences of high-dimensional graph data, and enables the identification and comparison of similar temporal states. We show the applicability of the technique to synthetic and real-world datasets, demonstrating that temporal patterns in dynamic graphs can be identified and interpreted over time. dg2pix contributes a suitable intermediate representation between node-link diagrams at the high detail end and matrix representations on the low detail end.

2022, Koh, Elliot, Blumenschein, Michael, Shao, Lin, Schreck, Tobias

Visualizing high-dimensional (HD) data is a key challenge for data scientists. The importance of this challenge is to properly map data properties, e.g., patterns, outliers, and correlations, from a HD data space onto a visualization. Parallel coordinate plots (PCPs) are a common way to do this. However, a PCP visualization can be arranged in several ways by reordering its axes, which may lead to different visual representations. Many methods have been developed with the aim of evaluating the quality of reorderings of given PCP view. A high-dimensional data set can be divided into multiple classes, and being able to identify differences between the classes is important. Then, besides overlaying the groups in a single PCP, we can show the different groups in individual PCPs in a small multiple fashion. This raises the problem of jointly reordering sets of PCPs to create meaningful reorderings of the set of plots. We propose a joint reordering strategy, based on maximizing the pairwise visual difference in PCPs, such as to support their contrastive comparison. We present an implementation and an evaluation of the reordering strategy to assess the effectiveness of the method. The approach shows feasible in bringing out pairwise difference in PCP plots and hence support comparison of grouped data.

2021, Beauxis-Aussalet, Emma, Behrisch, Michael, Borgo, Rita, Chau, Duen Horng, Collins, Christopher, El-Assady, Mennatallah, Keim, Daniel A., Oelke, Daniela, Schreck, Tobias, Strobelt, Hendrik

The increasing use of artificial intelligence (AI) technologies across application domains has prompted our society to pay closer attention to AI's trustworthiness, fairness, interpretability, and accountability. In order to foster trust in AI, it is important to consider the potential of interactive visualization, and how such visualizations help build trust in AI systems. This manifesto discusses the relevance of interactive visualizations and makes the following four claims: i) trust is not a technical problem, ii) trust is dynamic, iii) visualization cannot address all aspects of trust, and iv) visualization is crucial for human agency in AI.

2022, Cakmak, Eren, Jäckle, Dominik, Schreck, Tobias, Keim, Daniel A., Fuchs, Johannes

Multiscale visualizations are typically used to analyze multiscale processes and data in various application domains, such as the visual exploration of hierarchical genome structures in molecular biology. However, creating such multiscale visualizations remains challenging due to the plethora of existing work and the expression ambiguity in visualization research. Up to today, there has been little work to compare and categorize multiscale visualizations to understand their design practices. In this work, we present a structured literature analysis to provide an overview of common design practices in multiscale visualization research. We systematically reviewed and categorized 122 published journal or conference papers between 1995 and 2020. We organized the reviewed papers in a taxonomy that reveals common design factors. Researchers and practitioners can use our taxonomy to explore existing work to create new multiscale navigation and visualization techniques. Based on the reviewed papers, we examine research trends and highlight open research challenges.

2020, Cakmak, Eren, Schäfer, Hanna, Buchmüller, Juri F., Fuchs, Johannes, Schreck, Tobias, Jordan, Alex, Keim, Daniel A.

Domain experts for collective animal behavior analyze relationships between single animal movers and groups of animalsover time and space to detect emergent group properties. A common way to interpret this type of data is to visualize it as aspatio-temporal network. Collective behavior data sets are often large, and may hence result in dense and highly connectednode-link diagrams, resulting in issues of node-overlap and edge clutter. In this design study, in an iterative design process, wedeveloped glyphs as a design for seamlessly encoding relationships and movement characteristics of a single mover or clustersof movers. Based on these glyph designs, we developed a visual exploration prototype, MotionGlyphs, that supports domainexperts in interactively filtering, clustering, and animating spatio-temporal networks for collective animal behavior analysis. Bymeans of an expert evaluation, we show how MotionGlyphs supports important tasks and analysis goals of our domain experts,and we give evidence of the usefulness for analyzing spatio-temporal networks of collective animal behavior.