Recent studies have reported a shortcoming of visual exploratory data analysis (EDA) that can lead analysts to report spurious findings. These findings have prompted the advocacy of incorporating belief elicitation within the data analysis process. However, the results from these studies primarily drew from laboratory experiments, which can differ from real-world analysis contexts. In this article, we present outcomes from a longitudinal study with students enrolled in a university-level visual analytics course tackling the VAST Challenge. The students formed teams that were randomly assigned to the belief elicitation and non-belief elicitation conditions. Our study results indicate teams that underwent belief elicitation adopted an intentional approach, while teams in the non-belief elicitation condition reported greater diversity in findings, aligning with prior research. Surprisingly, teams from both conditions achieved equal success in solving the VAST Challenge, suggesting that analysts can incorporate belief elicitation strategically for different goals. We provide guidelines for incorporating belief elicitation into data analysis and teaching material for educators to include belief elicitation in visual analytics courses.

Visual validation of regression models in scatterplots is a common practice for assessing model quality, yet its efficacy remains unquantified. We conducted two empirical experiments to investigate individuals' ability to visually validate linear regression models (linear trends) and to examine the impact of common visualization designs on validation quality. The first experiment showed that the level of accuracy for visual estimation of slope (i.e., fitting a line to data) is higher than for visual validation of s lope (i.e., accepting a shown line). Notably, we found bias toward slopes that are "too steep" in both cases. This lead to novel insights that participants naturally assessed regression with orthogonal distances between the points and the line (i.e., ODR regression) rather than the common vertical distances (OLS regression). In the second experiment, we investigated whether incorporating common designs for regression visualization (error lines, bounding boxes, and confidence intervals) would improve visual validation. Even though error lines reduced validation bias, results failed to show the desired improvements in accuracy for any design. Overall, our findings suggest caution in using visual model validation for linear trends in scatterplots.

Despite decision-making being a vital goal of data visualization, little work has been done to differentiate decision-making tasks within the field. While visualization task taxonomies and typologies exist, they often focus on more granular analytical tasks that are too low-level to describe large complex decisions, which can make it difficult to reason about and design decision-support tools. In this paper, we contribute a typology of decision-making tasks that were iteratively refined from a list of design goals distilled from a literature review. Our typology is concise and consists of only three tasks: CHOOSE, ACTIVATE, and CREATE. Although decision types originating in other disciplines exist, we provide definitions for these tasks that are suitable for the visualization community. Our proposed typology offers two benefits. First, the ability to compose and hierarchically organize the tasks enables flexible and clear descriptions of decisions with varying levels of complexities. Second, the typology encourages productive discourse between visualization designers and domain experts by abstracting the intricacies of data, thereby promoting clarity and rigorous analysis of decision-making processes. We demonstrate the benefits of our typology through four case studies, and present an evaluation of the typology from semi-structured interviews with experienced members of the visualization community who have contributed to developing or publishing decision support systems for domain experts. Our interviewees used our typology to delineate the decision-making processes supported by their systems, demonstrating its descriptive capacity and effectiveness. Finally, we present preliminary findings on the usefulness of our typology for visualization design.

Dimensionality reduction techniques are widely used for visualizing high-dimensional data. However, support for interpreting patterns of dimension reduction results in the context of the original data space is often insufficient. Consequently, users may struggle to extract insights from the projections. In this paper, we introduce DimBridge, a visual analytics tool that allows users to interact with visual patterns in a projection and retrieve corresponding data patterns. DimBridge supports several interactions, allowing users to perform various analyses, from contrasting multiple clusters to explaining complex latent structures. Leveraging first-order predicate logic, DimBridge identifies subspaces in the original dimensions relevant to a queried pattern and provides an interface for users to visualize and interact with them. We demonstrate how DimBridge can help users overcome the challenges associated with interpreting visual patterns in projections.

We are excited to welcome you to IEEE VIS 2024 in sunny St. Pete Beach, Florida! The conference program is shaping up to be one of the best we have seen, and the conference venue is undoubtedly one of the most fun locations we have ever held the VIS conference.

Dashboards serve as effective visualization tools for conveying complex information. However, there exists a knowledge gap regarding how dashboard designs impact user engagement, necessitating designers to rely on their design expertise. Saliency has been used to comprehend viewing behaviors and assess visualizations, yet existing saliency models are primarily designed for single-view visualizations. To address this, we conduct an eye-tracking study to quantify participants' viewing patterns on dashboards. We collect eye-movement data from 60 participants, each viewing 36 dashboards (16 representative dashboards shared across all and 20 unique to each participant), totaling 1,216 dashboards and 2,160 eye-movement data instances. Analysis of the data from 16 dashboards viewed by all participants provides insights into how dashboard objects and layout designs influence viewing behaviors. Our analysis confirms known viewing patterns and reveals new patterns related to dashboard layout designs. Using the eye-movement data and identified patterns, we develop a saliency model to predict viewing behaviors with dashboards. Compared to state-of-the-art models for single-view visualizations, our model demonstrates overall improvement in prediction performance for dashboards. Finally, we propose potential dashboard design guidelines, illustrate an application case, and discuss general scanning strategies along with limitations and future work.

Data integration is often performed to consolidate information from multiple disparate data sources during visual data analysis. However, integration operations are usually separate from visual analytics operations such as encode and filter in both interface design and empirical research. We conducted a preliminary user study to investigate whether and how data integration should be incorporated directly into the visual analytics process. We used two interface alternatives featuring contrasting approaches to the data preparation and analysis workflow: manual file-based ex-situ integration as a separate step from visual analytics operations; and automatic UI-based in-situ integration merged with visual analytics operations. Participants were asked to complete specific and free-form tasks with each interface, browsing for patterns, generating insights, and summarizing relationships between attributes distributed across multiple files. Analyzing participants' interactions and feedback, we found both task completion time and total interactions to be similar across interfaces and tasks, as well as unique integration strategies between interfaces and emergent behaviors related to satisficing and cognitive bias. Participants' time spent and interactions revealed that in-situ integration enabled users to spend more time on analysis tasks compared with ex-situ integration. Participants' integration strategies and analytical behaviors revealed differences in interface usage for generating and tracking hypotheses and insights. With these results, we synthesized preliminary guidelines for designing future visual analytics interfaces that can support integrating attributes throughout an active analysis process.

This study presents insights from interviews with nineteen Knowledge Graph (KG) practitioners who work in both enterprise and academic settings on a wide variety of use cases. Through this study, we identify critical challenges experienced by KG practitioners when creating, exploring, and analyzing KGs that could be alleviated through visualization design. Our findings reveal three major personas among KG practitioners - KG Builders, Analysts, and Consumers - each of whom have their own distinct expertise and needs. We discover that KG Builders would benefit from schema enforcers, while KG Analysts need customizable query builders that provide interim query results. For KG Consumers, we identify a lack of efficacy for node-link diagrams, and the need for tailored domain-specific visualizations to promote KG adoption and comprehension. Lastly, we find that implementing KGs effectively in practice requires both technical and social solutions that are not addressed with current tools, technologies, and collaborative workflows. From the analysis of our interviews, we distill several visualization research directions to improve KG usability, including knowledge cards that balance digestibility and discoverability, timeline views to track temporal changes, interfaces that support organic discovery, and semantic explanations for AI and machine learning predictions.

Presenting a predictive model's performance is a communication bottleneck that threatens collaborations between data scientists and subject matter experts. Accuracy and error metrics alone fail to tell the whole story of a model - its risks, strengths, and limitations - making it difficult for subject matter experts to feel confident in their decision to use a model. As a result, models may fail in unexpected ways or go entirely unused, as subject matter experts disregard poorly presented models in favor of familiar, yet arguably substandard methods. In this paper, we describe an iterative study conducted with both subject matter experts and data scientists to understand the gaps in communication between these two groups. We find that, while the two groups share common goals of understanding the data and predictions of the model, friction can stem from unfamiliar terms, metrics, and visualizations - limiting the transfer of knowledge to SMEs and discouraging clarifying questions being asked during presentations. Based on our findings, we derive a set of communication guidelines that use visualization as a common medium for communicating the strengths and weaknesses of a model. We provide a demonstration of our guidelines in a regression modeling scenario and elicit feedback on their use from subject matter experts. From our demonstration, subject matter experts were more comfortable discussing a model's performance, more aware of the trade-offs for the presented model, and better equipped to assess the model's risks - ultimately informing and contextualizing the model's use beyond text and numbers.

Multiple-view (MV) visualizations have become ubiquitous for visual communication and exploratory data visualization. However, most existing MV visualizations are designed for the desktop, which can be unsuitable for the continuously evolving displays of varying screen sizes. In this article, we present a two-stage adaptation framework that supports the automated retargeting and semi-automated tailoring of a desktop MV visualization for rendering on devices with displays of varying sizes. First, we cast layout retargeting as an optimization problem and propose a simulated annealing technique that can automatically preserve the layout of multiple views. Second, we enable fine-tuning for the visual appearance of each view, using a rule-based auto configuration method complemented with an interactive interface for chart-oriented encoding adjustment. To demonstrate the feasibility and expressivity of our proposed approach, we present a gallery of MV visualizations that have been adapted from the desktop to small displays. We also report the result of a user study comparing visualizations generated using our approach with those by existing methods. The outcome indicates that the participants generally prefer visualizations generated using our approach and find them to be easier to use.

Projection techniques are often used to visualize high-dimensional data, allowing users to better understand the overall structure of multi-dimensional spaces on a 2D screen. Although many such methods exist, comparably little work has been done on generalizable methods of inverse-projection - the process of mapping the projected points, or more generally, the projection space back to the original high-dimensional space. In this article we present NNInv, a deep learning technique with the ability to approximate the inverse of any projection or mapping. NNInv learns to reconstruct high-dimensional data from any arbitrary point on a 2D projection space, giving users the ability to interact with the learned high-dimensional representation in a visual analytics system. We provide an analysis of the parameter space of NNInv, and offer guidance in selecting these parameters. We extend validation of the effectiveness of NNInv through a series of quantitative and qualitative analyses. We then demonstrate the method's utility by applying it to three visualization tasks: interactive instance interpolation, classifier agreement, and gradient visualization.

Abstract Projection algorithms such as t‐SNE or UMAP are useful for the visualization of high dimensional data, but depend on hyperparameters which must be tuned carefully. Unfortunately, iteratively recomputing projections to find the optimal hyperparameter values is computationally intensive and unintuitive due to the stochastic nature of such methods. In this paper we propose HyperNP, a scalable method that allows for real‐time interactive hyperparameter exploration of projection methods by training neural network approximations. A HyperNP model can be trained on a fraction of the total data instances and hyperparameter configurations that one would like to investigate and can compute projections for new data and hyperparameters at interactive speeds. HyperNP models are compact in size and fast to compute, thus allowing them to be embedded in lightweight visualization systems. We evaluate the performance of HyperNP across three datasets in terms of performance and speed. The results suggest that HyperNP models are accurate, scalable, interactive, and appropriate for use in real‐world settings.

Progressive visualization is fast becoming a technique in the visualization community to help users interact with large amounts of data. With progressive visualization, users can examine intermediate results of complex or long running computations, without waiting for the computation to complete. While this has shown to be beneficial to users, recent research has identified potential risks. For example, users may misjudge the uncertainty in the intermediate results and draw incorrect conclusions or see patterns that are not present in the final results. In this article, we conduct a comprehensive set of studies to quantify the advantages and limitations of progressive visualization. Based on a recent report by Micallef et al., we examine four types of cognitive biases that can occur with progressive visualization: uncertainty bias, illusion bias, control bias, and anchoring bias. The results of the studies suggest a cautious but promising use of progressive visualization - while there can be significant savings in task completion time, accuracy can be negatively affected in certain conditions. These findings confirm earlier reports of the benefits and drawbacks of progressive visualization and that continued research into mitigating the effects of cognitive biases is necessary.

Interactive visualization design and research have primarily focused on local data and synchronous events. However, for more complex use cases-e.g., remote database access and streaming data sources-developers must grapple with distributed data and asynchronous events. Currently, constructing these use cases is difficult and time-consuming; developers are forced to operationally program low-level details like asynchronous database querying and reactive event handling. This approach is in stark contrast to modern methods for browser-based interactive visualization, which feature high-level declarative specifications. In response, we present DIEL, a declarative framework that supports asynchronous events over distributed data. As in many declarative languages, DIEL developers specify only what data they want, rather than procedural steps for how to assemble it. Uniquely, DIEL models asynchronous events (e.g., user interactions, server responses) as streams of data that are captured in event logs. To specify the state of a visualization at any time, developers write declarative queries over the data and event logs; DIEL compiles and optimizes a corresponding dataflow graph, and automatically generates necessary low-level distributed systems details. We demonstrate DIEL'S performance and expressivity through example interactive visualizations that make diverse use of remote data and asynchronous events. We further evaluate DIEL'S usability using the Cognitive Dimensions of Notations framework, revealing wins such as ease of change, and compromises such as premature commitments.

Visual exploration of large multi-dimensional datasets has seen tremendous progress in recent years, allowing users to express rich data queries that produce informative visual summaries, all in real time. Techniques based on data cubes are some of the most promising approaches. However, these techniques usually require a large memory footprint for large datasets. To tackle this problem, we present NeuralCubes: neural networks that predict results for aggregate queries, similar to data cubes. NeuralCubes learns a function that takes as input a given query, for instance, a geographic region and temporal interval, and outputs the result of the query. The learned function serves as a real-time, low-memory approximator for aggregation queries. Our models are small enough to be sent to the client side (e.g. the web browser for a web-based application) for evaluation, enabling data exploration of large datasets without database/network connection. We demonstrate the effectiveness of NeuralCubes through extensive experiments on a variety of datasets and discuss how NeuralCubes opens up opportunities for new types of visualization and interaction.

Interaction enables users to navigate large amounts of data effectively, supports cognitive processing, and increases data representation methods. However, there have been few attempts to empirically demonstrate whether adding interaction to a static visualization improves its function beyond popular beliefs. In this paper, we address this gap. We use a classic Bayesian reasoning task as a testbed for evaluating whether allowing users to interact with a static visualization can improve their reasoning. Through two crowdsourced studies, we show that adding interaction to a static Bayesian reasoning visualization does not improve participants’ accuracy on a Bayesian reasoning task. In some cases, it can significantly detract from it. Moreover, we demonstrate that underlying visualization design modulates performance and that people with high versus low spatial ability respond differently to different interaction techniques and underlying base visualizations. Our work suggests that interaction is not as unambiguously good as we often believe; a well designed static visualization can be as, if not more, effective than an interactive one.

research-article Share on Special Issue on Interactive Visual Analytics for Making Explainable and Accountable Decisions Authors: Cagatay Turkay University of Warwick, Coventry, UK University of Warwick, Coventry, UKView Profile , Tatiana Von Landesberger University of Cologne and University of Rostock, Cologne, Germany University of Cologne and University of Rostock, Cologne, GermanyView Profile , Daniel Archambault Swansea University, Swansea, Wales, UK Swansea University, Swansea, Wales, UKView Profile , Shixia Liu Tsinghua University, Beijing, People’s Republic of China Tsinghua University, Beijing, People’s Republic of ChinaView Profile , Remco Chang Tufts University, Medford, USA Tufts University, Medford, USAView Profile Authors Info & Claims ACM Transactions on Interactive Intelligent SystemsVolume 11Issue 3-4December 2021 Article No.: 17pp 1–4https://doi.org/10.1145/3471903Online:03 September 2021Publication History 0citation187DownloadsMetricsTotal Citations0Total Downloads187Last 12 Months187Last 6 weeks20 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access

Most visual analytics systems assume that all foraging for data happens before the analytics process; once analysis begins, the set of data attributes considered is fixed. Such separation of data construction from analysis precludes iteration that can enable foraging informed by the needs that arise in-situ during the analysis. The separation of the foraging loop from the data analysis tasks can limit the pace and scope of analysis. In this paper, we present CAVA, a system that integrates data curation and data augmentation with the traditional data exploration and analysis tasks, enabling information foraging in-situ during analysis. Identifying attributes to add to the dataset is difficult because it requires human knowledge to determine which available attributes will be helpful for the ensuing analytical tasks. CAVA crawls knowledge graphs to provide users with a a broad set of attributes drawn from external data to choose from. Users can then specify complex operations on knowledge graphs to construct additional attributes. CAVA shows how visual analytics can help users forage for attributes by letting users visually explore the set of available data, and by serving as an interface for query construction. It also provides visualizations of the knowledge graph itself to help users understand complex joins such as multi-hop aggregations. We assess the ability of our system to enable users to perform complex data combinations without programming in a user study over two datasets. We then demonstrate the generalizability of CAVA through two additional usage scenarios. The results of the evaluation confirm that CAVA is effective in helping the user perform data foraging that leads to improved analysis outcomes, and offer evidence in support of integrating data augmentation as a part of the visual analytics pipeline.

Multiple-view visualization (MV) is a layout design technique often employed to help users see a large number of data attributes and values in a single cohesive representation. Because of its generalizability, the MV design has been widely adopted by the visualization community to help users examine and interact with large, complex, and high-dimensional data. However, although ubiquitous, there has been little work to categorize and analyze MVs in order to better understand its design space. As a result, there has been little to no guideline in how to use the MV design effectively. In this paper, we present an in-depth study of how MVs are designed in practice. We focus on two fundamental measures of multiple-view patterns: composition, which quantifies what view types and how many are there; and configuration, which characterizes spatial arrangement of view layouts in the display space. We build a new dataset containing 360 images of MVs collected from IEEE VIS, EuroVis, and PacificVis publications 2011 to 2019, and make fine-grained annotations of view types and layouts for these visualization images. From this data we conduct composition and configuration analyses using quantitative metrics of term frequency and layout topology. We identify common practices around MVs, including relationship of view types, popular view layouts, and correlation between view types and layouts. We combine the findings into a MV recommendation system, providing interactive tools to explore the design space, and support example-based design.

Static scatterplots often suffer from the overdraw problem on big datasets where object overlap causes undesirable visual clutter. The use of zooming in scatterplots can help alleviate this problem. With multiple zoom levels, more screen real estate is available, allowing objects to be placed in a less crowded way. We call this type of visualization scalable scatterplot visualizations, or SSV for short. Despite the potential of SSVs, existing systems and toolkits fall short in supporting the authoring of SSVs due to three limitations. First, many systems have limited scalability, assuming that data fits in the memory of one computer. Second, too much developer work, e.g., using custom code to generate mark layouts or render objects, is required. Third, many systems focus on only a small subset of the SSV design space (e.g. supporting a specific type of visual marks). To address these limitations, we have developed Kyrix-S, a system for easy authoring of SSVs at scale. Kyrix-S derives a declarative grammar that enables specification of a variety of SSVs in a few tens of lines of code, based on an existing survey of scatterplot tasks and designs. The declarative grammar is supported by a distributed layout algorithm which automatically places visual marks onto zoom levels. We store data in a multi-node database and use multi-node spatial indexes to achieve interactive browsing of large SSVs. Extensive experiments show that 1) Kyrix-S enables interactive browsing of SSVs of billions of objects, with response times under 500ms and 2) Kyrix-S achieves 4X-9X reduction in specification compared to a state-of-the-art authoring system.

Recent visual analytics systems make use of multiple machine learning models to better fit the data as opposed to traditional single, pre-defined model systems. However, while multi-model visual analytic systems can be effective, their added complexity adds usability concerns, as users are required to interact with the parameters of multiple models. Further, the advent of various model algorithms and associated hyperparameters creates an exhaustive model space to sample models from. This poses complexity to navigate this model space to find the right model for the data and the task. In this paper, we present Gaggle, a multi-model visual analytic system that enables users to interactively navigate the model space. Further translating user interactions into inferences, Gaggle simplifies working with multiple models by automatically finding the best model from the high-dimensional model space to support various user tasks. Through a qualitative user study, we show how our approach helps users to find a best model for a classification and ranking task. The study results confirm that Gaggle is intuitive and easy to use, supporting interactive model space navigation and automated model selection without requiring any technical expertise from users.

Abstract Building effective classifiers requires providing the modeling algorithms with information about the training data and modeling goals in order to create a model that makes proper tradeoffs. Machine learning algorithms allow for flexible specification of such meta‐information through the design of the objective functions that they solve. However, such objective functions are hard for users to specify as they are a specific mathematical formulation of their intents. In this paper, we present an approach that allows users to generate objective functions for classification problems through an interactive visual interface. Our approach adopts a semantic interaction design in that user interactions over data elements in the visualization are translated into objective function terms. The generated objective functions are solved by a machine learning solver that provides candidate models, which can be inspected by the user, and used to suggest refinements to the specifications. We demonstrate a visual analytics system QUESTO for users to manipulate objective functions to define domain‐specific constraints. Through a user study we show that QUESTO helps users create various objective functions that satisfy their goals.

Abstract There is fast‐growing literature on provenance‐related research, covering aspects such as its theoretical framework, use cases, and techniques for capturing, visualizing, and analyzing provenance data. As a result, there is an increasing need to identify and taxonomize the existing scholarship. Such an organization of the research landscape will provide a complete picture of the current state of inquiry and identify knowledge gaps or possible avenues for further investigation. In this STAR, we aim to produce a comprehensive survey of work in the data visualization and visual analytics field that focus on the analysis of user interaction and provenance data. We structure our survey around three primary questions: (1) WHY analyze provenance data, (2) WHAT provenance data to encode and how to encode it, and (3) HOW to analyze provenance data. A concluding discussion provides evidence‐based guidelines and highlights concrete opportunities for future development in this emerging area. The survey and papers discussed can be explored online interactively at https://provenance-survey.caleydo.org .

Latency in a visualization system is widely believed to affect user behavior in measurable ways, such as requiring the user to wait for the visualization system to respond, leading to interruption of the analytic flow. While this effect is frequently observed and widely accepted, precisely how latency affects different analysis scenarios is less well understood. In this paper, we examine the role of latency in the context of visual search, an essential task in data foraging and exploration using visualization. We conduct a series of studies on Amazon Mechanical Turk and find that under certain conditions, latency is a statistically significant predictor of visual search behavior, which is consistent with previous studies. However, our results also suggest that task type, task complexity, and other factors can modulate the effect of latency, in some cases rendering latency statistically insignificant in predicting user behavior. This suggests a more nuanced view of the role of latency than previously reported. Building on these results and the findings of prior studies, we propose design guidelines for measuring and interpreting the effects of latency when evaluating performance on visual search tasks.

The performance of deep learning models is dependent on the precise configuration of many layers and parameters. However, there are currently few systematic guidelines for how to configure a successful model. This means model builders often have to experiment with different configurations by manually programming different architectures (which is tedious and time consuming) or rely on purely automated approaches to generate and train the architectures (which is expensive). In this paper, we present Rapid Exploration of Model Architectures and Parameters, or REMAP, a visual analytics tool that allows a model builder to discover a deep learning model quickly via exploration and rapid experimentation of neural network architectures. In REMAP, the user explores the large and complex parameter space for neural network architectures using a combination of global inspection and local experimentation. Through a visual overview of a set of models, the user identifies interesting clusters of architectures. Based on their findings, the user can run ablation and variation experiments to identify the effects of adding, removing, or replacing layers in a given architecture and generate new models accordingly. They can also handcraft new models using a simple graphical interface. As a result, a model builder can build deep learning models quickly, efficiently, and without manual programming. We inform the design of REMAP through a design study with four deep learning model builders. Through a use case, we demonstrate that REMAP allows users to discover performant neural network architectures efficiently using visual exploration and user-defined semi-automated searches through the model space.

Abstract Many visual analytics systems allow users to interact with machine learning models towards the goals of data exploration and insight generation on a given dataset. However, in some situations, insights may be less important than the production of an accurate predictive model for future use. In that case, users are more interested in generating of diverse and robust predictive models, verifying their performance on holdout data, and selecting the most suitable model for their usage scenario. In this paper, we consider the concept of Exploratory Model Analysis (EMA), which is defined as the process of discovering and selecting relevant models that can be used to make predictions on a data source. We delineate the differences between EMA and the well‐known term exploratory data analysis in terms of the desired outcome of the analytic process: insights into the data or a set of deployable models. The contributions of this work are a visual analytics system workflow for EMA, a user study, and two use cases validating the effectiveness of the workflow. We found that our system workflow enabled users to generate complex models, to assess them for various qualities, and to select the most relevant model for their task.

Abstract Pan and zoom are basic yet powerful interaction techniques for exploring large datasets. However, existing zoomable UI toolkits such as Pad++ and ZVTM do not provide the backend database support and data‐driven primitives that are necessary for creating large‐scale visualizations. This limitation in existing general‐purpose toolkits has led to many purpose‐built solutions (e.g. Google Maps and ForeCache) that address the issue of scalability but cannot be easily extended to support visualizations beyond their intended data types and usage scenarios. In this paper, we introduce Kyrix to ease the process of creating general and large‐scale web‐based pan/zoom visualizations. Kyrix is an integrated system that provides the developer with a concise and expressive declarative language along with a backend support for performance optimization of large‐scale data. To evaluate the scalability of Kyrix, we conducted a set of benchmarked experiments and show that Kyrix can support high interactivity (with an average latency of 100 ms or below) on pan/zoom visualizations of 100 million data points. We further demonstrate the accessibility of Kyrix through an observational study with 8 developers. Results indicate that developers can quickly learn Kyrix's underlying declarative model to create scalable pan/zoom visualizations. Finally, we provide a gallery of visualizations and show that Kyrix is expressive and flexible in that it can support the developer in creating a wide range of customized visualizations across different application domains and data types.

In order to reduce the possibility of neural injury from seizures and sidestep the need for a neurologist to spend hours on manually reviewing the EEG recording, it is critical to automatically detect and classify "interictal-ictal continuum" (IIC) patterns from EEG data. However, the existing IIC classification techniques are shown to be not accurate and robust enough for clinical use because of the lack of high quality labels of EEG segments as training data. Obtaining high-quality labeled data is traditionally a manual process by trained clinicians that can be tedious, time-consuming, and error-prone. In this work, we propose Smile, an industrial scale system that provides an end-to-end solution to the IIC pattern classification problem. The core components of Smile include a visualization-based time series labeling module and a deep-learning based active learning module. The labeling module enables the users to explore and label 350 million EEG segments (30TB) at interactive speed. The multiple coordinated views allow the users to examine the EEG signals from both time domain and frequency domain simultaneously. The active learning module first trains a deep neural network that automatically extracts both the local features with respect to each segment itself and the long term dynamics of the EEG signals to classify IIC patterns. Then leveraging the output of the deep learning model, the EEG segments that can best improve the model are selected and prompted to clinicians to label. This process is iterated until the clinicians and the models show high degree of agreement. Our initial experimental results show that our Smile system allows the clinicians to label the EEG segments at will with a response time below 500 ms. The accuracy of the model is progressively improved as more and more high quality labels are acquired over time.

This January 2019 issue of theIEEE Transactions on Visualization and Computer Graphics (TVCG)contains the proceedings of IEEE VIS 2018, held during 21-26 October 2018 at the Estrel Hotel & Congress Center in Berlin. With IEEE VIS 2018, the conference series is in its 29th year.IEEE VIS consists of three conferences, held concurrently: the IEEE Visual Analytics Science and Technology Conference (IEEE VAST), the IEEE Information Visualization Conference (IEEE InfoVis), and the IEEE Scientific Visualization Conference (IEEE SciVis). These three conferences are the premier venues for the visualization community to exchange the latest ideas and developments, attracting researchers and practitioners alike.

When inspecting information visualizations under time critical settings, such as emergency response or monitoring the heart rate in a surgery room, the user only has a small amount of time to view the visualization "at a glance". In these settings, it is important to provide a quantitative measure of the visualization to understand whether or not the visualization is too "complex" to accurately judge at a glance. This paper proposes Pixel Approximate Entropy (PAE), which adapts the approximate entropy statistical measure commonly used to quantify regularity and unpredictability in time-series data, as a measure of visual complexity for line charts. We show that PAE is correlated with user-perceived chart complexity, and that increased chart PAE correlates with reduced judgement accuracy. 'We also find that the correlation between PAE values and participants' judgment increases when the user has less time to examine the line charts.

Recent visualization research efforts have incorporated experimental techniques and perceptual models from the vision science community. Perceptual laws such as Weber's law, for example, have been used to model the perception of correlation in scatterplots. While this thread of research has progressively refined the modeling of the perception of correlation in scatterplots, it remains unclear as to why such perception can be modeled using relatively simple functions, e.g., linear and log-linear. In this paper, we investigate a longstanding hypothesis that people use visual features in a chart as a proxy for statistical measures like correlation. For a given scatterplot, we extract 49 candidate visual features and evaluate which best align with existing models and participant judgments. The results support the hypothesis that people attend to a small number of visual features when discriminating correlation in scatterplots. We discuss how this result may account for prior conflicting findings, and how visual features provide a baseline for future model-based approaches in visualization evaluation and design.

"How common is interactive visualization on the web?" "What is the most popular visualization design?" "How prevalent are pie charts really?" These questions intimate the role of interactive visualization in the real (online) world. In this paper, we present our approach (and findings) to answering these questions. First, we introduce Beagle, which mines the web for SVG-based visualizations and automatically classifies them by type (i.e., bar, pie, etc.). With Beagle, we extract over 41,000 visualizations across five different tools and repositories, and classify them with 85% accuracy, across 24 visualization types. Given this visualization collection, we study usage across tools. We find that most visualizations fall under four types: bar charts, line charts, scatter charts, and geographic maps. Though controversial, pie charts are relatively rare for the visualization tools that were studied. Our findings also suggest that the total visualization types supported by a given tool could factor into its ease of use. However this effect appears to be mitigated by providing a variety of diverse expert visualization examples to users.

Prostate cancer is the most common cancer among men in the US, and yet most cases represent localized cancer for which the optimal treatment is unclear. Accumulating evidence suggests that the available treatment options, including surgery and conservative treatment, result in a similar prognosis for most men with localized prostate cancer. However, approximately 90% of patients choose surgery over conservative treatment, despite the risk of severe side effects like erectile dysfunction and incontinence. Recent medical research suggests that a key reason is the lack of patient-centered tools that can effectively communicate personalized risk information and enable them to make better health decisions. In this paper, we report the iterative design process and results of developing the PROgnosis Assessment for Conservative Treatment (PROACT) tool, a personalized health risk communication tool for localized prostate cancer patients. PROACT utilizes two published clinical prediction models to communicate the patients' personalized risk estimates and compare treatment options. In collaboration with the Maine Medical Center, we conducted two rounds of evaluations with prostate cancer survivors and urologists to identify the design elements and narrative structure that effectively facilitate patient comprehension under emotional distress. Our results indicate that visualization can be an effective means to communicate complex risk information to patients with low numeracy and visual literacy. However, the visualizations need to be carefully chosen to balance readability with ease of comprehension. In addition, due to patients' charged emotional state, an intuitive narrative structure that considers the patients' information need is critical to aid the patients' comprehension of their risk information.

We present Brain Automated Chorales (BACh), an adaptive brain-computer system that dynamically increases the levels of difficulty in a musical learning task based on pianists' cognitive workload measured by functional near-infrared spectroscopy. As users' cognitive workload fell below a certain threshold, suggesting that they had mastered the material and could handle more cognitive information, BACh automatically increased the difficulty of the learning task. We found that learners played with significantly increased accuracy and speed in the brain-based adaptive task compared to our control condition. Participant feedback indicated that they felt they learned better with BACh and they liked the timings of the level changes. The underlying premise of BACh can be applied to learning situations where a task can be broken down into increasing levels of difficulty.

In this paper, we present ForeCache, a general-purpose tool for exploratory browsing of large datasets. ForeCache utilizes a client-server architecture, where the user interacts with a lightweight client-side interface to browse datasets, and the data to be browsed is retrieved from a DBMS running on a back-end server. We assume a detail-on-demand browsing paradigm, and optimize the back-end support for this paradigm by inserting a separate middleware layer in front of the DBMS. To improve response times, the middleware layer fetches data ahead of the user as she explores a dataset.

Decades of research have repeatedly shown that people perform poorly at estimating and understanding conditional probabilities that are inherent in Bayesian reasoning problems. Yet in the medical domain, both physicians and patients make daily, life-critical judgments based on conditional probability. Although there have been a number of attempts to develop more effective ways to facilitate Bayesian reasoning, reports of these findings tend to be inconsistent and sometimes even contradictory. For instance, the reported accuracies for individuals being able to correctly estimate conditional probability range from 6% to 62%. In this work, we show that problem representation can significantly affect accuracies. By controlling the amount of information presented to the user, we demonstrate how text and visualization designs can increase overall accuracies to as high as 77%. Additionally, we found that for users with high spatial ability, our designs can further improve their accuracies to as high as 100%. By and large, our findings provide explanations for the inconsistent reports on accuracy in Bayesian reasoning tasks and show a significant improvement over existing methods. We believe that these findings can have immediate impact on risk communication in health-related fields.

Information graphics, or infographics, combine elements of data visualization with design and have become an increasingly popular means for disseminating data. While several studies have suggested that aesthetics in visualization and infographics relate to desirable outcomes like engagement and memorability, it remains unknown how quickly aesthetic impressions are formed, and what it is that makes an infographic appealing. We address these questions by analyzing 1,278 participants' ratings on appeal after seeing infographics for 500ms. Our results establish that: 1) people form a reliable first impression of the appeal of an infographic based on a mere exposure effect, 2) this first impression is largely based on colorfulness and visual complexity, and 3) age, gender, and education level influence the preferred level of colorfulness and complexity. More generally, these findings suggest that outcomes such as engagement and memorability might be determined much earlier than previously thought.

Individual differences matter. While this has been the theme for many recent works in the Visualization and HCI communities, the mystery of how to develop personalized visualizations remains. This is largely because very little is known about how users actually use visualizations to solve problems and even less is known about how individual differences affect these problem-solving strategies. In this paper, we provide evidence that strategies are indeed influenced by individual differences. We demonstrate how the personality trait locus of control impacts strategies on hierarchical visualizations, and we introduce design recommendations for personalized visualizations.

Dynamic difficulty adjustments can be used in human-computer systems in order to improve user engagement and performance. In this paper, we use functional near-infrared spectroscopy (fNIRS) to obtain passive brain sensing data and detect extended periods of boredom or overload. From these physiological signals, we can adapt a simulation in order to optimize workload in real-time, which allows the system to better fit the task to the user from moment to moment. To demonstrate this idea, we ran a laboratory study in which participants performed path planning for multiple unmanned aerial vehicles (UAVs) in a simulation. Based on their state, we varied the difficulty of the task by adding or removing UAVs and found that we were able to decrease error by 35% over a baseline condition. Our results show that we can use fNIRS brain sensing to detect task difficulty in real-time and construct an interface that improves user performance through dynamic difficulty adjustment.

The bubble cursor is a promising cursor expansion technique, improving a user's movement time and accuracy in pointing tasks. We introduce a brain-based target expansion system, which improves the efficacy of bubble cursor by increasing the expansion of high importance targets at the optimal time based on brain measurements correlated to a particular type of multitasking. We demonstrate through controlled experiments that brain-based target expansion can deliver a graded and continuous level of assistance to a user according to their cognitive state, thereby improving task and speed-accuracy metrics, even without explicit visual changes to the system. Such an adaptation is ideal for use in complex systems to steer users toward higher priority goals during times of increased demand.

This editorial introduction describes the aims and scope of ACM Transactions on Interactive Intelligent Systems 's special issue on interactive computational visual analytics. It explains why visual analytics is crucial to the growing needs surrounding data analysis, and it shows how the four articles selected for this issue reflect this theme.

Visual analytics is inherently a collaboration between human and computer. However, in current visual analytics systems, the computer has limited means of knowing about its users and their analysis processes. While existing research has shown that a user's interactions with a system reflect a large amount of the user's reasoning process, there has been limited advancement in developing automated, real-time techniques that mine interactions to learn about the user. In this paper, we demonstrate that we can accurately predict a user's task performance and infer some user personality traits by using machine learning techniques to analyze interaction data. Specifically, we conduct an experiment in which participants perform a visual search task, and apply well-known machine learning algorithms to three encodings of the users' interaction data. We achieve, depending on algorithm and encoding, between 62% and 83% accuracy at predicting whether each user will be fast or slow at completing the task. Beyond predicting performance, we demonstrate that using the same techniques, we can infer aspects of the user's personality factors, including locus of control, extraversion, and neuroticism. Further analyses show that strong results can be attained with limited observation time: in one case 95% of the final accuracy is gained after a quarter of the average task completion time. Overall, our findings show that interactions can provide information to the computer about its human collaborator, and establish a foundation for realizing mixed-initiative visual analytics systems.

Despite years of research yielding systems and guidelines to aid visualization design, practitioners still face the challenge of identifying the best visualization for a given dataset and task. One promising approach to circumvent this problem is to leverage perceptual laws to quantitatively evaluate the effectiveness of a visualization design. Following previously established methodologies, we conduct a large scale (n=1687) crowdsourced experiment to investigate whether the perception of correlation in nine commonly used visualizations can be modeled using Weber's law. The results of this experiment contribute to our understanding of information visualization by establishing that: (1) for all tested visualizations, the precision of correlation judgment could be modeled by Weber's law, (2) correlation judgment precision showed striking variation between negatively and positively correlated data, and (3) Weber models provide a concise means to quantify, compare, and rank the perceptual precision afforded by a visualization.

Agent-based simulation has become a key technique for modeling and simulating dynamic, complicated behaviors in social and behavioral sciences. As these simulations become more complex, they generate an increasingly large amount of data. Lacking the appropriate tools and support, it has become difficult for social scientists to interpret and analyze the results of these simulations. In this paper, we introduce the Aggregate Temporal Graph (ATG), a graph formulation that can be used to capture complex relationships between discrete simulation states in time. Using this formulation, we can assist social scientists in identifying critical simulation states by examining graph substructures. In particular, we define the concept of a Gateway and its inverse, a Terminal, which capture the relationships between pivotal states in the simulation and their inevitable outcomes. We propose two real-time computable algorithms to identify these relationships and provide a proof of correctness, complexity analysis, and empirical run-time analysis. We demonstrate the use of these algorithms on a large-scale social science simulation of political power and violence in present-day Thailand, and discuss broader applications of the ATG and associated algorithms in other domains such as analytic provenance.

Modern database management systems (DBMS) have been designed to efficiently store, manage and perform computations on massive amounts of data. In contrast, many existing visualization systems do not scale seamlessly from small data sets to enormous ones. We have designed a three-tiered visualization system called ScalaR to deal with this issue. ScalaR dynamically performs resolution reduction when the expected result of a DBMS query is too large to be effectively rendered on existing screen real estate. Instead of running the original query, ScalaR inserts aggregation, sampling or filtering operations to reduce the size of the result. This paper presents the design and implementation of ScalaR, and shows results for an example application, displaying satellite imagery data stored in SciDB as the back-end DBMS.

Recent research suggests that individual personality differences can influence performance with visualizations. In addition to stable personality traits, research in psychology has found that temporary changes in affect (emotion) can also significantly impact performance during cognitive tasks. In this paper, we show that affective priming also influences user performance on visual judgment tasks through an experiment that combines affective priming with longstanding graphical perception experiments. Our results suggest that affective priming can influence accuracy in common graphical perception tasks. We discuss possible explanations for these findings, and describe how these findings can be applied to design visualizations that are less (or more) susceptible to error in common visualization contexts.

We show how brain sensing can lend insight to the evaluation of visual interfaces and establish a role for fNIRS in visualization. Research suggests that the evaluation of visual design benefits by going beyond performance measures or questionnaires to measurements of the user's cognitive state. Unfortunately, objectively and unobtrusively monitoring the brain is difficult. While functional near-infrared spectroscopy (fNIRS) has emerged as a practical brain sensing technology in HCI, visual tasks often rely on the brain's quick, massively parallel visual system, which may be inaccessible to this measurement. It is unknown whether fNIRS can distinguish differences in cognitive state that derive from visual design alone. In this paper, we use the classic comparison of bar graphs and pie charts to test the viability of fNIRS for measuring the impact of a visual design on the brain. Our results demonstrate that we can indeed measure this impact, and furthermore measurements indicate that there are not universal differences in bar graphs and pie charts.

Existing research suggests that individual personality differences are correlated with a user's speed and accuracy in solving problems with different types of complex visualization systems. We extend this research by isolating factors in personality traits as well as in the visualizations that could have contributed to the observed correlation. We focus on a personality trait known as "locus of control” (LOC), which represents a person's tendency to see themselves as controlled by or in control of external events. To isolate variables of the visualization design, we control extraneous factors such as color, interaction, and labeling. We conduct a user study with four visualizations that gradually shift from a list metaphor to a containment metaphor and compare the participants' speed, accuracy, and preference with their locus of control and other personality factors. Our findings demonstrate that there is indeed a correlation between the two: participants with an internal locus of control perform more poorly with visualizations that employ a containment metaphor, while those with an external locus of control perform well with such visualizations. These results provide evidence for the externalization theory of visualization. Finally, we propose applications of these findings to adaptive visual analytics and visualization evaluation.

Visual Analytics is "the science of analytical reasoning facilitated by visual interactive interfaces". The goal of this field is to develop tools and methodologies for approaching problems whose size and complexity render them intractable without the close coupling of both human and machine analysis. Researchers have explored this coupling in many venues: VAST, Vis, InfoVis, CHI, KDD, IUI, and more. While there have been myriad promising examples of human-computer collaboration, there exists no common language for comparing systems or describing the benefits afforded by designing for such collaboration. We argue that this area would benefit significantly from consensus about the design attributes that define and distinguish existing techniques. In this work, we have reviewed 1,271 papers from many of the top-ranking conferences in visual analytics, human-computer interaction, and visualization. From these, we have identified 49 papers that are representative of the study of human-computer collaborative problem-solving, and provide a thorough overview of the current state-of-the-art. Our analysis has uncovered key patterns of design hinging on human and machine-intelligence affordances, and also indicates unexplored avenues in the study of this area. The results of this analysis provide a common framework for understanding these seemingly disparate branches of inquiry, which we hope will motivate future work in the field.

Urban environments require cognitive abilities focused on both spatial overview and detailed understanding of uses and places. These abilities are distinct but overlap and reinforce each other. Our work quantitatively and qualitatively measures the effects on a user's overall understanding of the environment after navigating with either a GPS or a road map in a previously unknown neighborhood. Experimental recall of spatial and semantic information indicates that using a road map enables subjects to demonstrate a significantly better spatial understanding, identify semantic elements more often using common terms, place semantic elements in spatial locations with greater accuracy and recall semantic elements in tighter clusters than when using a GPS. We conclude that a spatial understanding is a necessary framework for organizing semantic information that is useful for inferred tasks.

Abstract We present a probe‐based interface for the exploration of the results of a geospatial simulation of urban growth. Because our interface allows the user great freedom in how they choose to define regions‐of‐interest to examine and compare, the classic geospatial analytic issue known as the modifiable areal unit problem (MAUP) quickly arises. The user may delineate regions with unseen differences that can affect the fairness of the comparisons made between them. To alleviate this problem, our interface first alerts the user if it detects any potential unfairness between regions when they are selected for comparison. It then presents the dimensions with potential problematic outliers to the user for evaluation. Finally, it provides a number of semi‐automated tools to assist the user in correcting their regions' boundaries to minimize the inequalities they feel could significantly impact their comparisons.

Abstract Bridges deteriorate over their life cycles and require continuous maintenance to ensure their structural integrity, and in turn, the safety of the public. Maintaining bridges is a multi‐faceted operation that requires both domain knowledge and analytics techniques over large data sources. Although most existing bridge management systems (BMS) are very efficient at data storage, they are not as effective at providing analytical capabilities or as flexible at supporting different inspection technologies. In this paper, we present a visual analytics system that extends the capability of current BMSs. Based on a nation‐wide survey and our interviews with bridge managers, we designed our system to be customizable so that it can provide interactive exploration, information correlation, and domain‐oriented data analysis. When tested by bridge managers of the U.S. Department of Transportation, we validated that our system provides bridge managers with the necessary features for performing in‐depth analysis of bridges from a variety of perspectives that are in accordance to their typical workflow.

Abstract Principle Component Analysis (PCA) is a widely used mathematical technique in many fields for factor and trend analysis, dimension reduction, etc. However, it is often considered to be a “black box” operation whose results are difficult to interpret and sometimes counter‐intuitive to the user. In order to assist the user in better understanding and utilizing PCA, we have developed a system that visualizes the results of principal component analysis using multiple coordinated views and a rich set of user interactions. Our design philosophy is to support analysis of multivariate datasets through extensive interaction with the PCA output. To demonstrate the usefulness of our system, we performed a comparative user study with a known commercial system, SAS/INSIGHT's Interactive Data Exploration. Participants in our study solved a number of high‐level analysis tasks with each interface and rated the systems on ease of learning and usefulness. Based on the participants' accuracy, speed, and qualitative feedback, we observe that our system helps users to better understand relationships between the data and the calculated eigenspace, which allows the participants to more accurately analyze the data. User feedback suggests that the interactivity and transparency of our system are the key strengths of our approach.

We present an interactive framework for exploring space-time and form-function relationships in experimentally collected high-resolution biomechanical data sets. These data describe complex 3D motions (e.g. chewing, walking, flying) performed by animals and humans and captured via high-speed imaging technologies, such as biplane fluoroscopy. In analyzing these 3D biomechanical motions, interactive 3D visualizations are important, in particular, for supporting spatial analysis. However, as researchers in information visualization have pointed out, 2D visualizations can also be effective tools for multi-dimensional data analysis, especially for identifying trends over time. Our approach, therefore, combines techniques from both 3D and 2D visualizations. Specifically, it utilizes a multi-view visualization strategy including a small multiples view of motion sequences, a parallel coordinates view, and detailed 3D inspection views. The resulting framework follows an overview first, zoom and filter, then details-on-demand style of analysis, and it explicitly targets a limitation of current tools, namely, supporting analysis and comparison at the level of a collection of motions rather than sequential analysis of a single or small number of motions. Scientific motion collections appropriate for this style of analysis exist in clinical work in orthopedics and physical rehabilitation, in the study of functional morphology within evolutionary biology, and in other contexts. An application is described based on a collaboration with evolutionary biologists studying the mechanics of chewing motions in pigs. Interactive exploration of data describing a collection of more than one hundred experimentally captured pig chewing cycles is described.

Abstract Many previous approaches to detecting urban change from LIDAR point clouds interpolate the points into rasters, perform pixel‐based image processing to detect changes, and produce 2D images as output. We present a method of LIDAR change detection that maintains accuracy by only using the raw, irregularly spaced LIDAR points, and extracts relevant changes as individual 3D models. We then utilize these models, alongside existing GIS data, within an interactive application that allows the chronological exploration of the changes to an urban environment. A three‐tiered level‐of‐detail system maintains a scale‐appropriate, legible visual representation across the entire range of view scales, from individual changes such as buildings and trees, to groups of changes such as new residential developments, deforestation, and construction sites, and finally to larger regions such as neighborhoods and districts of a city that are emerging or undergoing revitalization. Tools are provided to assist the visual analysis by urban planners and historians through semantic categorization and filtering of the changes presented.

Abstract Recent increases in terrorist activity around the world have made analyzing and understanding such activities more critical than ever. With the help of organizations such as the National Center for the Study of Terrorism and Responses to Terrorism (START), we now have detailed historical information on each terrorist event around the world since 1970. However, due to the size and complexity of the data, identifying terrorists' patterns and trends has been difficult. To better enable investigators in understanding terrorist activities, we propose a visual analytical system that focuses on depicting one of the most fundamental concepts in investigative analysis, the five W's (who, what, where, when, and why). Views in our system are highly correlated, and each represents one of the W's. With this approach, an investigator can interactively explore terrorist activities efficiently and discover reasons of attacks (why) by identifying patterns temporally (when), geo‐spatially (where), between multiple terrorist groups (who), and across different methods or modes of attacks (what). By coupling a global perspective with the details gleaned from asking these five questions, the system allows analysts to think both tactically and strategically.

Traditional geospatial information visualizations often present views that restrict the user to a single perspective. When zoomed out, local trends and anomalies become suppressed and lost; when zoomed in for local inspection, spatial awareness and comparison between regions become limited. In our model, coordinated visualizations are integrated within individual probe interfaces, which depict the local data in user-defined regions-of-interest. Our probe concept can be incorporated into a variety of geospatial visualizations to empower users with the ability to observe, coordinate, and compare data across multiple local regions. It is especially useful when dealing with complex simulations or analyses where behavior in various localities differs from other localities and from the system as a whole. We illustrate the effectiveness of our technique over traditional interfaces by incorporating it within three existing geospatial visualization systems: an agent-based social simulation, a census data exploration tool, and an 3D GIS environment for analyzing urban change over time. In each case, the probe-based interaction enhances spatial awareness, improves inspection and comparison capabilities, expands the range of scopes, and facilitates collaboration among multiple users.

Numerous systems have been developed to display large collections of data for urban contexts; however, most have focused on layering of single dimensions of data and manual calculations to understand relationships within the urban environment. Furthermore, these systems often limit the userâs perspectives on the data, thereby diminishing the userâs spatial understanding of the viewing region. In this paper, we introduce a highly interactive urban visualization tool that provides intuitive understanding of the urban data. Our system utilizes an aggregation method that combines buildings and city blocks into legible clusters, thus providing continuous levels of abstraction while preserving the userâs mental model of the city. In conjunction with a 3D view of the urban model, a separate but integrated information visualization view displays multiple disparate dimensions of the urban data, allowing the user to understand the urban environment both spatially and cognitively in one glance. For our evaluation, expert users from various backgrounds viewed a real city model with census data and confirmed that our system allowed them to gain more intuitive and deeper understanding of the urban model from different perspectives and levels of abstraction than existing commercial urban visualization systems.