Interactive Visual Analysis of Structure-borne Noise Data.

Numerical simulation has become omnipresent in the automotive domain, posing new challenges such as high-dimensional parameter spaces and large as well as incomplete and multi-faceted data. In this design study, we show how interactive visual exploration and analysis of high-dimensional, spectral data from noise simulation can facilitate design improvements in the context of conflicting criteria. Here, we focus on structure-borne noise, i.e., noise from vibrating mechanical parts. Detecting problematic noise sources early in the design and production process is essential for reducing a product's development costs and its time to market. In a close collaboration of visualization and automotive engineering, we designed a new, interactive approach to quickly identify and analyze critical noise sources, also contributing to an improved understanding of the analyzed system. Several carefully designed, interactive linked views enable the exploration of noises, vibrations, and harshness at multiple levels of detail, both in the frequency and spatial domain. This enables swift and smooth changes of perspective; selections in the frequency domain are immediately reflected in the spatial domain, and vice versa. Noise sources are quickly identified and shown in the context of their neighborhood, both in the frequency and spatial domain. We propose a novel drill-down view, especially tailored to noise data analysis. Split boxplots and synchronized 3D geometry views support comparison tasks. With this solution, engineers iterate over design optimizations much faster, while maintaining a good overview at each iteration. We evaluated the new approach in the automotive industry, studying noise simulation data for an internal combustion engine.

On Sketch-Based Selections From Scatterplots Using KDE, Compared to Mahalanobis and CNN Brushing.

Fast and accurate brushing is crucial in visual data exploration and sketch-based solutions are successful methods. In this article, we detail a solution, based on kernel density estimation, which computes a data subset selection in a scatterplot from a simple click-and-drag interaction. We explain how this technique relates to two alternative approaches, i.e., Mahalanobis brushing and CNN brushing. To study this relation, we conducted two user studies and present both a quantitative three-fold comparison as well as additional details about the prevalence of all possible cases in that each technique succeeds/fails. With this, we also provide a comparison between empirical modeling and implicit modeling by DL in terms of accuracy, efficiency, generality, and interpretability.

Integrated Dual Analysis of Quantitative and Qualitative High-Dimensional Data.

The Dual Analysis framework is a powerful enabling technology for the exploration of high dimensional quantitative data by treating data dimensions as first-class objects that can be explored in tandem with data values. In this article, we extend the Dual Analysis framework through the joint treatment of quantitative (numerical) and qualitative (categorical) dimensions. Computing common measures for all dimensions allows us to visualize both quantitative and qualitative dimensions in the same view. This enables a natural joint treatment of mixed data during interactive visual exploration and analysis. Several measures of variation for nominal qualitative data can also be applied to ordinal qualitative and quantitative data. For example, instead of measuring variability from a mean or median, other measures assess inter-data variation or average variation from a mode. In this work, we demonstrate how these measures can be integrated into the Dual Analysis framework to explore and generate hypotheses about high-dimensional mixed data. A medical case study using clinical routine data of patients suffering from Cerebral Small Vessel Disease (CSVD), conducted with a senior neurologist and a medical student, shows that a joint Dual Analysis approach for quantitative and qualitative data can rapidly lead to new insights based on which new hypotheses may be generated.

DimLift: Interactive Hierarchical Data Exploration Through Dimensional Bundling.

The identification of interesting patterns and relationships is essential to exploratory data analysis. This becomes increasingly difficult in high dimensional datasets. While dimensionality reduction techniques can be utilized to reduce the analysis space, these may unintentionally bury key dimensions within a larger grouping and obfuscate meaningful patterns. With this work we introduce DimLift, a novel visual analysis method for creating and interacting with dimensional bundles. Generated through an iterative dimensionality reduction or user-driven approach, dimensional bundles are expressive groups of dimensions that contribute similarly to the variance of a dataset. Interactive exploration and reconstruction methods via a layered parallel coordinates plot allow users to lift interesting and subtle relationships to the surface, even in complex scenarios of missing and mixed data types. We exemplify the power of this technique in an expert case study on clinical cohort data alongside two additional case examples from nutrition and ecology.

Sketch-Based Fast and Accurate Querying of Time Series Using Parameter-Sharing LSTM Networks.

Sketching is one common approach to query time series data for patterns of interest. Most existing solutions for matching the data with the interaction are based on an empirically modeled similarity function between the user's sketch and the time series data with limited efficiency and accuracy. In this article, we introduce a machine learning based solution for fast and accurate querying of time series data based on a swift sketching interaction. We build on existing LSTM technology (long short-term memory) to encode both the sketch and the time series data in a network with shared parameters. We use data from a user study to let the network learn a proper similarity function. We focus our approach on perceived similarities and achieve that the learned model also includes a user-side aspect. To the best of our knowledge, this is the first data-driven solution for querying time series data in visual analytics. Besides evaluating the accuracy and efficiency directly in a quantitative way, we also compare our solution to the recently published Qetch algorithm as well as the commonly used dynamic time warping (DTW) algorithm.

IsoTrotter: Visually Guided Empirical Modelling of Atmospheric Convection.

Empirical models, fitted to data from observations, are often used in natural sciences to describe physical behaviour and support discoveries. However, with more complex models, the regression of parameters quickly becomes insufficient, requiring a visual parameter space analysis to understand and optimize the models. In this work, we present a design study for building a model describing atmospheric convection. We present a mixed-initiative approach to visually guided modelling, integrating an interactive visual parameter space analysis with partial automatic parameter optimization. Our approach includes a new, semi-automatic technique called IsoTrotting, where we optimize the procedure by navigating along isocontours of the model. We evaluate the model with unique observational data of atmospheric convection based on flight trajectories of paragliders.

Scale-Space Splatting: Reforming Spacetime for Cross-Scale Exploration of Integral Measures in Molecular Dynamics.

Understanding large amounts of spatiotemporal data from particle-based simulations, such as molecular dynamics, often relies on the computation and analysis of aggregate measures. These, however, by virtue of aggregation, hide structural information about the space/time localization of the studied phenomena. This leads to degenerate cases where the measures fail to capture distinct behaviour. In order to drill into these aggregate values, we propose a multi-scale visual exploration technique. Our novel representation, based on partial domain aggregation, enables the construction of a continuous scale-space for discrete datasets and the simultaneous exploration of scales in both space and time. We link these two scale-spaces in a scale-space space-time cube and model linked views as orthogonal slices through this cube, thus enabling the rapid identification of spatio-temporal patterns at multiple scales. To demonstrate the effectiveness of our approach, we showcase an advanced exploration of a protein-ligand simulation.

Multiscale Visual Drilldown for the Analysis of Large Ensembles of Multi-Body Protein Complexes.

When studying multi-body protein complexes, biochemists use computational tools that can suggest hundreds or thousands of their possible spatial configurations. However, it is not feasible to experimentally verify more than only a very small subset of them. In this paper, we propose a novel multiscale visual drilldown approach that was designed in tight collaboration with proteomic experts, enabling a systematic exploration of the configuration space. Our approach takes advantage of the hierarchical structure of the data - from the whole ensemble of protein complex configurations to the individual configurations, their contact interfaces, and the interacting amino acids. Our new solution is based on interactively linked 2D and 3D views for individual hierarchy levels. At each level, we offer a set of selection and filtering operations that enable the user to narrow down the number of configurations that need to be manually scrutinized. Furthermore, we offer a dedicated filter interface, which provides the users with an overview of the applied filtering operations and enables them to examine their impact on the explored ensemble. This way, we maintain the history of the exploration process and thus enable the user to return to an earlier point of the exploration. We demonstrate the effectiveness of our approach on two case studies conducted by collaborating proteomic experts.

Personalized Sketch-Based Brushing in Scatterplots.

Brushing is at the heart of most modern visual analytics solutions and effective and efficient brushing is crucial for successful interactive data exploration and analysis. As the user plays a central role in brushing, several data-driven brushing tools have been designed that are based on predicting the user's brushing goal. All of these general brushing models learn the users' average brushing preference, which is not optimal for every single user. In this paper, we propose an innovative framework that offers the user opportunities to improve the brushing technique while using it. We realized this framework with a CNN-based brushing technique and the result shows that with additional data from a particular user, the model can be refined (better performance in terms of accuracy), eventually converging to a personalized model based on a moderate amount of retraining.

Comparative visualization of protein secondary structures.

BACKGROUND: Protein function is determined by many factors, namely by its constitution, spatial arrangement, and dynamic behavior. Studying these factors helps the biochemists and biologists to better understand the protein behavior and to design proteins with modified properties. One of the most common approaches to these studies is to compare the protein structure with other molecules and to reveal similarities and differences in their polypeptide chains. RESULTS: We support the comparison process by proposing a new visualization technique that bridges the gap between traditionally used 1D and 3D representations. By introducing the information about mutual positions of protein chains into the 1D sequential representation the users are able to observe the spatial differences between the proteins without any occlusion commonly present in 3D view. Our representation is designed to serve namely for comparison of multiple proteins or a set of time steps of molecular dynamics simulation. CONCLUSIONS: The novel representation is demonstrated on two usage scenarios. The first scenario aims to compare a set of proteins from the family of cytochromes P450 where the position of the secondary structures has a significant impact on the substrate channeling. The second scenario focuses on the protein flexibility when by comparing a set of time steps our representation helps to reveal the most dynamically changing parts of the protein chain.

Interactive exploration of ligand transportation through protein tunnels.

BACKGROUND: Protein structures and their interaction with ligands have been in the focus of biochemistry and structural biology research for decades. The transportation of ligand into the protein active site is often complex process, driven by geometric and physico-chemical properties, which renders the ligand path full of jitter and impasses. This prevents understanding of the ligand transportation and reasoning behind its behavior along the path. RESULTS: To address the needs of the domain experts we design an explorative visualization solution based on a multi-scale simplification model. It helps to navigate the user to the most interesting parts of the ligand trajectory by exploring different attributes of the ligand and its movement, such as its distance to the active site, changes of amino acids lining the ligand, or ligand "stuckness". The process is supported by three linked views - 3D representation of the simplified trajectory, scatterplot matrix, and bar charts with line representation of ligand-lining amino acids. CONCLUSIONS: The usage of our tool is demonstrated on molecular dynamics simulations provided by the domain experts. The tool was tested by the domain experts from protein engineering and the results confirm that it helps to navigate the user to the most interesting parts of the ligand trajectory and to understand the ligand behavior.

A Fractional Cartesian Composition Model for Semi-Spatial Comparative Visualization Design.

The study of spatial data ensembles leads to substantial visualization challenges in a variety of applications. In this paper, we present a model for comparative visualization that supports the design of according ensemble visualization solutions by partial automation. We focus on applications, where the user is interested in preserving selected spatial data characteristics of the data as much as possible-even when many ensemble members should be jointly studied using comparative visualization. In our model, we separate the design challenge into a minimal set of user-specified parameters and an optimization component for the automatic configuration of the remaining design variables. We provide an illustrated formal description of our model and exemplify our approach in the context of several application examples from different domains in order to demonstrate its generality within the class of comparative visualization problems for spatial data ensembles.

Designing Progressive and Interactive Analytics Processes for High-Dimensional Data Analysis.

In interactive data analysis processes, the dialogue between the human and the computer is the enabling mechanism that can lead to actionable observations about the phenomena being investigated. It is of paramount importance that this dialogue is not interrupted by slow computational mechanisms that do not consider any known temporal human-computer interaction characteristics that prioritize the perceptual and cognitive capabilities of the users. In cases where the analysis involves an integrated computational method, for instance to reduce the dimensionality of the data or to perform clustering, such non-optimal processes are often likely. To remedy this, progressive computations, where results are iteratively improved, are getting increasing interest in visual analytics. In this paper, we present techniques and design considerations to incorporate progressive methods within interactive analysis processes that involve high-dimensional data. We define methodologies to facilitate processes that adhere to the perceptual characteristics of users and describe how online algorithms can be incorporated within these. A set of design recommendations and according methods to support analysts in accomplishing high-dimensional data analysis tasks are then presented. Our arguments and decisions here are informed by observations gathered over a series of analysis sessions with analysts from finance. We document observations and recommendations from this study and present evidence on how our approach contribute to the efficiency and productivity of interactive visual analysis sessions involving high-dimensional data.

Interactively illustrating polymerization using three-level model fusion.

BACKGROUND: Research in cell biology is steadily contributing new knowledge about many aspects of physiological processes, both with respect to the involved molecular structures as well as their related function. Illustrations of the spatio-temporal development of such processes are not only used in biomedical education, but also can serve scientists as an additional platform for in-silico experiments. RESULTS: In this paper, we contribute a new, three-level modeling approach to illustrate physiological processes from the class of polymerization at different time scales. We integrate physical and empirical modeling, according to which approach best suits the different involved levels of detail, and we additionally enable a form of interactive steering, while the process is illustrated. We demonstrate the suitability of our approach in the context of several polymerization processes and report from a first evaluation with domain experts. CONCLUSION: We conclude that our approach provides a new, hybrid modeling approach for illustrating the process of emergence in physiology, embedded in a densely filled environment. Our approach of a complementary fusion of three systems combines the strong points from the different modeling approaches and is capable to bridge different spatial and temporal scales.

Interactive visual analysis of heterogeneous cohort-study data.

Medical cohort studies enable the study of medical hypotheses with many samples. Often, these studies acquire a large amount of heterogeneous data from many subjects. Usually, researchers study a specific data subset to confirm or reject specific hypotheses. A new approach enables the interactive visual exploration and analysis of such data, helping to generate and validate hypotheses. A data-cube-based model handles partially overlapping data subsets during the interactive visualization. This model enables seamless integration of the heterogeneous data and the linking of spatial and nonspatial views of the data. Researchers implemented this model in a prototype application and used it to analyze data acquired in a cohort study on cognitive aging. Case studies employed the prototype to study aspects of brain connectivity, demonstrating the model's potential and flexibility.

Characterizing cancer subtypes using dual analysis in Caleydo StratomeX.

Dual analysis uses statistics to describe both the dimensions and rows of a high-dimensional dataset. Researchers have integrated it into StratomeX, a Caleydo view for cancer subtype analysis. In addition, significant-difference plots show the elements of a candidate subtype that differ significantly from other subtypes, thus letting analysts characterize subtypes. Analysts can also investigate how data samples relate to their assigned subtype and other groups. This approach lets them create well-defined subtypes based on statistical properties. Three case studies demonstrate the approach's utility, showing how it reproduced findings from a published subtype characterization.

Visual Methods for Analyzing Probabilistic Classification Data.

Multi-class classifiers often compute scores for the classification samples describing probabilities to belong to different classes. In order to improve the performance of such classifiers, machine learning experts need to analyze classification results for a large number of labeled samples to find possible reasons for incorrect classification. Confusion matrices are widely used for this purpose. However, they provide no information about classification scores and features computed for the samples. We propose a set of integrated visual methods for analyzing the performance of probabilistic classifiers. Our methods provide insight into different aspects of the classification results for a large number of samples. One visualization emphasizes at which probabilities these samples were classified and how these probabilities correlate with classification error in terms of false positives and false negatives. Another view emphasizes the features of these samples and ranks them by their separation power between selected true and false classifications. We demonstrate the insight gained using our technique in a benchmarking classification dataset, and show how it enables improving classification performance by interactively defining and evaluating post-classification rules.

Visual Analytics for Complex Engineering Systems: Hybrid Visual Steering of Simulation Ensembles.

In this paper we propose a novel approach to hybrid visual steering of simulation ensembles. A simulation ensemble is a collection of simulation runs of the same simulation model using different sets of control parameters. Complex engineering systems have very large parameter spaces so a naïve sampling can result in prohibitively large simulation ensembles. Interactive steering of simulation ensembles provides the means to select relevant points in a multi-dimensional parameter space (design of experiment). Interactive steering efficiently reduces the number of simulation runs needed by coupling simulation and visualization and allowing a user to request new simulations on the fly. As system complexity grows, a pure interactive solution is not always sufficient. The new approach of hybrid steering combines interactive visual steering with automatic optimization. Hybrid steering allows a domain expert to interactively (in a visualization) select data points in an iterative manner, approximate the values in a continuous region of the simulation space (by regression) and automatically find the "best" points in this continuous region based on the specified constraints and objectives (by optimization). We argue that with the full spectrum of optimization options, the steering process can be improved substantially. We describe an integrated system consisting of a simulation, a visualization, and an optimization component. We also describe typical tasks and propose an interactive analysis workflow for complex engineering systems. We demonstrate our approach on a case study from automotive industry, the optimization of a hydraulic circuit in a high pressure common rail Diesel injection system.

Attribute Signatures: Dynamic Visual Summaries for Analyzing Multivariate Geographical Data.

The visual analysis of geographically referenced datasets with a large number of attributes is challenging due to the fact that the characteristics of the attributes are highly dependent upon the locations at which they are focussed, and the scale and time at which they are measured. Specialized interactive visual methods are required to help analysts in understanding the characteristics of the attributes when these multiple aspects are considered concurrently. Here, we develop attribute signatures-interactively crafted graphics that show the geographic variability of statistics of attributes through which the extent of dependency between the attributes and geography can be visually explored. We compute a number of statistical measures, which can also account for variations in time and scale, and use them as a basis for our visualizations. We then employ different graphical configurations to show and compare both continuous and discrete variation of location and scale. Our methods allow variation in multiple statistical summaries of multiple attributes to be considered concurrently and geographically, as evidenced by examples in which the census geography of London and the wider UK are explored.

Radial sets: interactive visual analysis of large overlapping sets.

In many applications, data tables contain multi-valued attributes that often store the memberships of the table entities to multiple sets such as which languages a person masters, which skills an applicant documents, or which features a product comes with. With a growing number of entities, the resulting element-set membership matrix becomes very rich of information about how these sets overlap. Many analysis tasks targeted at set-typed data are concerned with these overlaps as salient features of such data. This paper presents Radial Sets, a novel visual technique to analyze set memberships for a large number of elements. Our technique uses frequency-based representations to enable quickly finding and analyzing different kinds of overlaps between the sets, and relating these overlaps to other attributes of the table entities. Furthermore, it enables various interactions to select elements of interest, find out if they are over-represented in specific sets or overlaps, and if they exhibit a different distribution for a specific attribute compared to the rest of the elements. These interactions allow formulating highly-expressive visual queries on the elements in terms of their set memberships and attribute values. As we demonstrate via two usage scenarios, Radial Sets enable revealing and analyzing a multitude of overlapping patterns between large sets, beyond the limits of state-of-the-art techniques.