Search

Inspired by the great success of machine learning (ML), researchers have applied ML techniques to visualizations to achieve a better design, development, and evaluation of visualizations. This branch of studies, known as ML4VIS, is gaining increasing research attention in recent years. To successfully adapt ML techniques for visualizations, a structured understanding of the integration of ML4VIS is needed. In this article, we systematically survey 88 ML4VIS studies, aiming to answer two motivating questions: “what visualization processes can be assisted by ML?” and “how ML techniques can be used to solve visualization problems? ”This survey reveals seven main processes where the employment of ML techniques can benefit visualizations: Data Processing4VIS, Data-VIS Mapping, Insight Communication, Style Imitation, VIS Interaction, VIS Reading, and User Profiling. The seven processes are related to existing visualization theoretical models in an ML4VIS pipeline, aiming to illuminate the role of ML-assisted visualization in general visualizations. Meanwhile, the seven processes are mapped into main learning tasks in ML to align the capabilities of ML with the needs in visualization. Current practices and future opportunities of ML4VIS are discussed in the context of the ML4VIS pipeline and the ML-VIS mapping. While more studies are still needed in the area of ML4VIS, we hope this article can provide a stepping-stone for future exploration. A web-based interactive browser of this survey is available at https://ml4vis.github.io. [ABSTRACT FROM AUTHOR]

In social networks, individuals’ decisions are strongly influenced by recommendations from their friends, acquaintances, and favorite renowned personalities. The popularity of online social networking platforms makes them the prime venues to advertise products and promote opinions. The Influence Maximization (IM) problem entails selecting a seed set of users that maximizes the influence spread, i.e., the expected number of users positively influenced by a stochastic diffusion process triggered by the seeds. Engineering and analyzing IM algorithms remains a difficult and demanding task due to the NP-hardness of the problem and the stochastic nature of the diffusion processes. Despite several heuristics being introduced, they often fail in providing enough information on how the network topology affects the diffusion process, precious insights that could help researchers improve their seed set selection. In this paper, we present VAIM, a visual analytics system that supports users in analyzing, evaluating, and comparing information diffusion processes determined by different IM algorithms. Furthermore, VAIM provides useful insights that the analyst can use to modify the seed set of an IM algorithm, so to improve its influence spread. We assess our system by: $(i)$ (i) a qualitative evaluation based on a guided experiment with two domain experts on two different data sets; $(ii)$ (i i) a quantitative estimation of the value of the proposed visualization through the ICE-T methodology by Wall et al. (IEEE TVCG - 2018). The twofold assessment indicates that VAIM effectively supports our target users in the visual analysis of the performance of IM algorithms. [ABSTRACT FROM AUTHOR]

Vision-based deep learning (DL) methods have made great progress in learning autonomous driving models from large-scale crowd-sourced video datasets. They are trained to predict instantaneous driving behaviors from video data captured by on-vehicle cameras. In this paper, we develop a geo-context aware visualization system for the study of Autonomous Driving Model (ADM) predictions together with large-scale ADM video data. The visual study is seamlessly integrated with the geographical environment by combining DL model performance with geospatial visualization techniques. Model performance measures can be studied together with a set of geospatial attributes over map views. Users can also discover and compare prediction behaviors of multiple DL models in both city-wide and street-level analysis, together with road images and video contents. Therefore, the system provides a new visual exploration platform for DL model designers in autonomous driving. Use cases and domain expert evaluation show the utility and effectiveness of the visualization system. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM PUBLISHER]

For effective flow visualization, identifying representative flow lines or surfaces is an important problem which has been studied. However, no work can solve the problem for both lines and surfaces. In this paper, we present FlowNet, a single deep learning framework for clustering and selection of streamlines and stream surfaces. Given a collection of streamlines or stream surfaces generated from a flow field data set, our approach converts them into binary volumes and then employs an autoencoder to learn their respective latent feature descriptors. These descriptors are used to reconstruct binary volumes for error estimation and network training. Once converged, the feature descriptors can well represent flow lines or surfaces in the latent space. We perform dimensionality reduction of these feature descriptors and cluster the projection results accordingly. This leads to a visual interface for exploring the collection of flow lines or surfaces via clustering, filtering, and selection of representatives. Intuitive user interactions are provided for visual reasoning of the collection with ease. We validate and explain our deep learning framework from multiple perspectives, demonstrate the effectiveness of FlowNet using several flow field data sets of different characteristics, and compare our approach against state-of-the-art streamline and stream surface selection algorithms. [ABSTRACT FROM AUTHOR]

The rapid development of Convolutional Neural Networks (CNNs) in recent years has triggered significant breakthroughs in many machine learning (ML) applications. The ability to understand and compare various CNN models available is thus essential. The conventional approach with visualizing each model's quantitative features, such as classification accuracy and computational complexity, is not sufficient for a deeper understanding and comparison of the behaviors of different models. Moreover, most of the existing tools for assessing CNN behaviors only support comparison between two models and lack the flexibility of customizing the analysis tasks according to user needs. This paper presents a visual analytics system, VAC-CNN (Visual Analytics for Comparing CNNs), that supports the in-depth inspection of a single CNN model as well as comparative studies of two or more models. The ability to compare a larger number of (e.g., tens of) models especially distinguishes our system from previous ones. With a carefully designed model visualization and explaining support, VAC-CNN facilitates a highly interactive workflow that promptly presents both quantitative and qualitative information at each analysis stage. We demonstrate VAC-CNN's effectiveness for assisting novice ML practitioners in evaluating and comparing multiple CNN models through two use cases and one preliminary evaluation study using the image classification tasks on the ImageNet dataset. [ABSTRACT FROM AUTHOR]

Personal informatics research helps people track personal data for the purposes of self-reflection and gaining self-knowledge. This field, however, has predominantly focused on the data collection and insight-generation elements of self-tracking, with less attention paid to flexible data analysis. As a result, this inattention has led to inflexible analytic pipelines that do not reflect or support the diverse ways people want to engage with their data. This paper contributes a review of personal informatics and visualization research literature to expose a gap in our knowledge for designing flexible tools that assist people engaging with and analyzing personal data in personal contexts, what we call the personal informatics analysis gap. We explore this gap through a multistage longitudinal study on how asthmatics engage with personal air quality data, and we report how participants: were motivated by broad and diverse goals; exhibited patterns in the way they explored their data; engaged with their data in playful ways; discovered new insights through serendipitous exploration; and were reluctant to use analysis tools on their own. These results present new opportunities for visual analysis research and suggest the need for fundamental shifts in how and what we design when supporting personal data analysis. [ABSTRACT FROM AUTHOR]

Graph mining is an essential component of recommender systems and search engines. Outputs of graph mining models typically provide a ranked list sorted by each item's relevance or utility. However, recent research has identified issues of algorithmic bias in such models, and new graph mining algorithms have been proposed to correct for bias. As such, algorithm developers need tools that can help them uncover potential biases in their models while also exploring the impacts of correcting for biases when employing fairness-aware algorithms. In this paper, we present FairRankVis, a visual analytics framework designed to enable the exploration of multi-class bias in graph mining algorithms. We support both group and individual fairness levels of comparison. Our framework is designed to enable model developers to compare multi-class fairness between algorithms (for example, comparing PageRank with a debiased PageRank algorithm) to assess the impacts of algorithmic debiasing with respect to group and individual fairness. We demonstrate our framework through two usage scenarios inspecting algorithmic fairness. [ABSTRACT FROM AUTHOR]

In the process of developing an infrastructure for providing visualization and visual analytics (VIS) tools to epidemiologists and modeling scientists, we encountered a technical challenge for applying a number of visual designs to numerous datasets rapidly and reliably with limited development resources. In this paper, we present a technical solution to address this challenge. Operationally, we separate the tasks of data management, visual designs, and plots and dashboard deployment in order to streamline the development workflow. Technically, we utilize: an ontology to bring datasets, visual designs, and deployable plots and dashboards under the same management framework; multi-criteria search and ranking algorithms for discovering potential datasets that match a visual design; and a purposely-design user interface for propagating each visual design to appropriate datasets (often in tens and hundreds) and quality-assuring the propagation before the deployment. This technical solution has been used in the development of the RAMPVIS infrastructure for supporting a consortium of epidemiologists and modeling scientists through visualization. [ABSTRACT FROM AUTHOR]

Visual analytics enables us to analyze huge information spaces in order to support complex decision making and data exploration. Humans play a central role in generating knowledge from the snippets of evidence emerging from visual data analysis. Although prior research provides frameworks that generalize this process, their scope is often narrowly focused so they do not encompass different perspectives at different levels. This paper proposes a knowledge generation model for visual analytics that ties together these diverse frameworks, yet retains previously developed models (e.g., KDD process) to describe individual segments of the overall visual analytic processes. To test its utility, a real world visual analytics system is compared against the model, demonstrating that the knowledge generation process model provides a useful guideline when developing and evaluating such systems. The model is used to effectively compare different data analysis systems. Furthermore, the model provides a common language and description of visual analytic processes, which can be used for communication between researchers. At the end, our model reflects areas of research that future researchers can embark on. [ABSTRACT FROM PUBLISHER]

An introduction to the papers presented at the visualization conferences organized by the Institute of Electrical & Electronics Engineers (IEEE) that was held from October 1-6, 2017 is presented which includes visual analytics, data manipulation, and the applications of visualization.

Deep learning methods are being increasingly used for urban traffic prediction where spatiotemporal traffic data is aggregated into sequentially organized matrices that are then fed into convolution-based residual neural networks. However, the widely known modifiable areal unit problem within such aggregation processes can lead to perturbations in the network inputs. This issue can significantly destabilize the feature embeddings and the predictions - rendering deep networks much less useful for the experts. This paper approaches this challenge by leveraging unit visualization techniques that enable the investigation of many-to-many relationships between dynamically varied multi-scalar aggregations of urban traffic data and neural network predictions. Through regular exchanges with a domain expert, we design and develop a visual analytics solution that integrates 1) a Bivariate Map equipped with an advanced bivariate colormap to simultaneously depict input traffic and prediction errors across space, 2) a Moran's I Scatterplot that provides local indicators of spatial association analysis, and 3) a Multi-scale Attribution View that arranges non-linear dot plots in a tree layout to promote model analysis and comparison across scales. We evaluate our approach through a series of case studies involving a real-world dataset of Shenzhen taxi trips, and through interviews with domain experts. We observe that geographical scale variations have important impact on prediction performances, and interactive visual exploration of dynamically varying inputs and outputs benefit experts in the development of deep traffic prediction models. [ABSTRACT FROM AUTHOR]

This paper introduces Polyphorm, an interactive visualization and model fitting tool that provides a novel approach for investigating cosmological datasets. Through a fast computational simulation method inspired by the behavior of Physarum polycephalum, an unicellular slime mold organism that efficiently forages for nutrients, astrophysicists are able to extrapolate from sparse datasets, such as galaxy maps archived in the Sloan Digital Sky Survey, and then use these extrapolations to inform analyses of a wide range of other data, such as spectroscopic observations captured by the Hubble Space Telescope. Researchers can interactively update the simulation by adjusting model parameters, and then investigate the resulting visual output to form hypotheses about the data. We describe details of Polyphorm's simulation model and its interaction and visualization modalities, and we evaluate Polyphorm through three scientific use cases that demonstrate the effectiveness of our approach. [ABSTRACT FROM AUTHOR]

In this paper, we introduce 11–20 (Image Insight 2020), a multimedia analytics approach for analytic categorization of image collections. Advanced visualizations for image collections exist, but they need tight integration with a machine model to support the task of analytic categorization. Directly employing computer vision and interactive learning techniques gravitates towards search. Analytic categorization, however, is not machine classification (the difference between the two is called the pragmatic gap): a human adds/redefines/deletes categories of relevance on the fly to build insight, whereas the machine classifier is rigid and non-adaptive. Analytic categorization that truly brings the user to insight requires a flexible machine model that allows dynamic sliding on the exploration-search axis, as well as semantic interactions: a human thinks about image data mostly in semantic terms. 11–20 brings three major contributions to multimedia analytics on image collections and towards closing the pragmatic gap. Firstly, a new machine model that closely follows the user's interactions and dynamically models her categories of relevance. II-20's machine model, in addition to matching and exceeding the state of the art's ability to produce relevant suggestions, allows the user to dynamically slide on the exploration-search axis without any additional input from her side. Secondly, the dynamic, 1-image-at-a-time Tetris metaphor that synergizes with the model. It allows a well-trained model to analyze the collection by itself with minimal interaction from the user and complements the classic grid metaphor. Thirdly, the fast-forward interaction, allowing the user to harness the model to quickly expand (“fast-forward”) the categories of relevance, expands the multimedia analytics semantic interaction dictionary. Automated experiments show that II-20's machine model outperforms the existing state of the art and also demonstrate the Tetris metaphor's analytic quality. User studies further confirm that II–20 is an intuitive, efficient, and effective multimedia analytics tool. [ABSTRACT FROM AUTHOR]

Analyzing interaction data provides an opportunity to learn about users, uncover their underlying goals, and create intelligent visualization systems. The first step for intelligent response in visualizations is to enable computers to infer user goals and strategies through observing their interactions with a system. Researchers have proposed multiple techniques to model users, however, their frameworks often depend on the visualization design, interaction space, and dataset. Due to these dependencies, many techniques do not provide a general algorithmic solution to user exploration modeling. In this paper, we construct a series of models based on the dataset and pose user exploration modeling as a Bayesian model selection problem where we maintain a belief over numerous competing models that could explain user interactions. Each of these competing models represent an exploration strategy the user could adopt during a session. The goal of our technique is to make high-level and in-depth inferences about the user by observing their low-level interactions. Although our proposed idea is applicable to various probabilistic model spaces, we demonstrate a specific instance of encoding exploration patterns as competing models to infer information relevance. We validate our technique's ability to infer exploration bias, predict future interactions, and summarize an analytic session using user study datasets. Our results indicate that depending on the application, our method outperforms established baselines for bias detection and future interaction prediction. Finally, we discuss future research directions based on our proposed modeling paradigm and suggest how practitioners can use this method to build intelligent visualization systems that understand users' goals and adapt to improve the exploration process. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) have demonstrated extraordinarily good performance in many computer vision tasks. The increasing size of CNN models, however, prevents them from being widely deployed to devices with limited computational resources, e.g., mobile/embedded devices. The emerging topic of model pruning strives to address this problem by removing less important neurons and fine-tuning the pruned networks to minimize the accuracy loss. Nevertheless, existing automated pruning solutions often rely on a numerical threshold of the pruning criteria, lacking the flexibility to optimally balance the trade-off between efficiency and accuracy. Moreover, the complicated interplay between the stages of neuron pruning and model fine-tuning makes this process opaque, and therefore becomes difficult to optimize. In this paper, we address these challenges through a visual analytics approach, named CNNPruner. It considers the importance of convolutional filters through both instability and sensitivity, and allows users to interactively create pruning plans according to a desired goal on model size or accuracy. Also, CNNPruner integrates state-of-the-art filter visualization techniques to help users understand the roles that different filters played and refine their pruning plans. Through comprehensive case studies on CNNs with real-world sizes, we validate the effectiveness of CNNPruner. [ABSTRACT FROM AUTHOR]

Causality is crucial to understanding the mechanisms behind complex systems and making decisions that lead to intended outcomes. Event sequence data is widely collected from many real-world processes, such as electronic health records, web clickstreams, and financial transactions, which transmit a great deal of information reflecting the causal relations among event types. Unfortunately, recovering causalities from observational event sequences is challenging, as the heterogeneous and high-dimensional event variables are often connected to rather complex underlying event excitation mechanisms that are hard to infer from limited observations. Many existing automated causal analysis techniques suffer from poor explainability and fail to include an adequate amount of human knowledge. In this paper, we introduce a visual analytics method for recovering causalities in event sequence data. We extend the Granger causality analysis algorithm on Hawkes processes to incorporate user feedback into causal model refinement. The visualization system includes an interactive causal analysis framework that supports bottom-up causal exploration, iterative causal verification and refinement, and causal comparison through a set of novel visualizations and interactions. We report two forms of evaluation: a quantitative evaluation of the model improvements resulting from the user-feedback mechanism, and a qualitative evaluation through case studies in different application domains to demonstrate the usefulness of the system. [ABSTRACT FROM AUTHOR]

In this paper, we present a visual analytics tool for enabling hypothesis-based evaluation of machine learning (ML) models. We describe a novel ML-testing framework that combines the traditional statistical hypothesis testing (commonly used in empirical research) with logical reasoning about the conclusions of multiple hypotheses. The framework defines a controlled configuration for testing a number of hypotheses as to whether and how some extra information about a “concept” or “feature” may benefit or hinder an ML model. Because reasoning multiple hypotheses is not always straightforward, we provide HypoML as a visual analysis tool, with which, the multi-thread testing results are first transformed to analytical results using statistical and logical inferences, and then to a visual representation for rapid observation of the conclusions and the logical flow between the testing results and hypotheses. We have applied HypoML to a number of hypothesized concepts, demonstrating the intuitive and explainable nature of the visual analysis. [ABSTRACT FROM AUTHOR]

Investigating relationships between variables in multi-dimensional data sets is a common task for data analysts and engineers. More specifically, it is often valuable to understand which ranges of which input variables lead to particular values of a given target variable. Unfortunately, with an increasing number of independent variables, this process may become cumbersome and time-consuming due to the many possible combinations that have to be explored. In this paper, we propose a novel approach to visualize correlations between input variables and a target output variable that scales to hundreds of variables. We developed a visual model based on neural networks that can be explored in a guided way to help analysts find and understand such correlations. First, we train a neural network to predict the target from the input variables. Then, we visualize the inner workings of the resulting model to help understand relations within the data set. We further introduce a new regularization term for the backpropagation algorithm that encourages the neural network to learn representations that are easier to interpret visually. We apply our method to artificial and real-world data sets to show its utility. [ABSTRACT FROM AUTHOR]

We propose a new semantic-level crowd evaluation metric in this paper. Crowd simulation has been an active and important area for several decades. However, only recently has there been an increased focus on evaluating the fidelity of the results with respect to real-world situations. The focus to date has been on analyzing the properties of low-level features such as pedestrian trajectories, or global features such as crowd densities. We propose the first approach based on finding semantic information represented by latent Path Patterns in both real and simulated data in order to analyze and compare them. Unsupervised clustering by non-parametric Bayesian inference is used to learn the patterns, which themselves provide a rich visualization of the crowd behavior. To this end, we present a new Stochastic Variational Dual Hierarchical Dirichlet Process ( SV-DHDP) model. The fidelity of the patterns is computed with respect to a reference, thus allowing the outputs of different algorithms to be compared with each other and/or with real data accordingly. Detailed evaluations and comparisons with existing metrics show that our method is a good alternative for comparing crowd data at a different level and also works with more types of data, holds fewer assumptions and is more robust to noise. [ABSTRACT FROM PUBLISHER]

In this paper we present a method for predicting the rendering time to display multi-dimensional data for the analysis of computer simulations using the HyperSlice [36] method with Gaussian process model reconstruction. Our method relies on a theoretical understanding of how the data points are drawn on slices and then fits the formula to a user’s machine using practical experiments. We also describe the typical characteristics of data when analyzing deterministic computer simulations as described by the statistics community. We then show the advantage of carefully considering how many data points can be drawn in real time by proposing two approaches of how this predictive formula can be used in a real-world system. [ABSTRACT FROM AUTHOR]

In this paper, we present an abstract model of visualization and inference processes, and describe an information-theoretic measure for optimizing such processes. In order to obtain such an abstraction, we first examined six classes of workflows in data analysis and visualization, and identified four levels of typical visualization components, namely disseminative, observational, analytical and model-developmental visualization. We noticed a common phenomenon at different levels of visualization, that is, the transformation of data spaces (referred to as alphabets) usually corresponds to the reduction of maximal entropy along a workflow. Based on this observation, we establish an information-theoretic measure of cost-benefit ratio that may be used as a cost function for optimizing a data visualization process. To demonstrate the validity of this measure, we examined a number of successful visualization processes in the literature, and showed that the information-theoretic measure can mathematically explain the advantages of such processes over possible alternatives. [ABSTRACT FROM PUBLISHER]

Tracking the temporal evolution of features in time-varying data is a key method in visualization. For typical feature definitions, such as vortices, objects are sparsely distributed over the data domain. In this paper, we present a novel approach for tracking both sparse and space-filling features. While the former comprise only a small fraction of the domain, the latter form a set of objects whose union covers the domain entirely while the individual objects are mutually disjunct. Our approach determines the assignment of features between two successive time-steps by solving two graph optimization problems. It first resolves one-to-one assignments of features by computing a maximum-weight, maximum-cardinality matching on a weighted bi-partite graph. Second, our algorithm detects events by creating a graph of potentially conflicting event explanations and finding a weighted, independent set in it. We demonstrate our method's effectiveness on synthetic and simulation data sets, the former of which enables quantitative evaluation because of the availability of ground-truth information. Here, our method performs on par or better than a well-established reference algorithm. In addition, manual visual inspection by our collaborators confirm the results’ plausibility for simulation data. [ABSTRACT FROM AUTHOR]

Social media, such as Facebook and WeChat, empowers millions of users to create, consume, and disseminate online information on an unprecedented scale. The abundant information on social media intensifies the competition of WeChat Public Official Articles (i.e., posts) for gaining user attention due to the zero-sum nature of attention. Therefore, only a small portion of information tends to become extremely popular while the rest remains unnoticed or quickly disappears. Such a typical “long-tail” phenomenon is very common in social media. Thus, recent years have witnessed a growing interest in predicting the future trend in the popularity of social media posts and understanding the factors that influence the popularity of the posts. Nevertheless, existing predictive models either rely on cumbersome feature engineering or sophisticated parameter tuning, which are difficult to understand and improve. In this paper, we study and enhance a point process-based model by incorporating visual reasoning to support communication between the users and the predictive model for a better prediction result. The proposed system supports users to uncover the working mechanism behind the model and improve the prediction accuracy accordingly based on the insights gained. We use realistic WeChat articles to demonstrate the effectiveness of the system and verify the improved model on a large scale of WeChat articles. We also elicit and summarize the feedback from WeChat domain experts. [ABSTRACT FROM AUTHOR]

Experts in data and physical sciences have to regularly grapple with the problem of competing models. Be it analytical or physics-based models, a cross-cutting challenge for experts is to reliably diagnose which model outcomes appropriately predict or simulate real-world phenomena. Expert judgment involves reconciling information across many, and often, conflicting criteria that describe the quality of model outcomes. In this paper, through a design study with climate scientists, we develop a deeper understanding of the problem and solution space of model diagnostics, resulting in the following contributions: i) a problem and task characterization using which we map experts' model diagnostics goals to multi-way visual comparison tasks, ii) a design space of comparative visual cues for letting experts quickly understand the degree of disagreement among competing models and gauge the degree of stability of model outputs with respect to alternative criteria, and iii) design and evaluation of MyriadCues, an interactive visualization interface for exploring alternative hypotheses and insights about good and bad models by leveraging comparative visual cues. We present case studies and subjective feedback by experts, which validate how MyriadCues enables more transparent model diagnostic mechanisms, as compared to the state of the art. [ABSTRACT FROM AUTHOR]

User behaviour analytics (UBA) systems offer sophisticated models that capture users' behaviour over time with an aim to identify fraudulent activities that do not match their profiles. Motivated by the challenges in the interpretation of UBA models, this paper presents a visual analytics approach to help analysts gain a comprehensive understanding of user behaviour at multiple levels, namely individual and group level. We take a user-centred approach to design a visual analytics framework supporting the analysis of collections of users and the numerous sessions of activities they conduct within digital applications. The framework is centred around the concept of hierarchical user profiles that are built based on features derived from sessions, as well as on user tasks extracted using a topic modelling approach to summarise and stratify user behaviour. We externalise a series of analysis goals and tasks, and evaluate our methods through use cases conducted with experts. We observe that with the aid of interactive visual hierarchical user profiles, analysts are able to conduct exploratory and investigative analysis effectively, and able to understand the characteristics of user behaviour to make informed decisions whilst evaluating suspicious users and activities. [ABSTRACT FROM AUTHOR]

Deep Neural Networks (DNNs) have been extensively used in multiple disciplines due to their superior performance. However, in most cases, DNNs are considered as black-boxes and the interpretation of their internal working mechanism is usually challenging. Given that model trust is often built on the understanding of how a model works, the interpretation of DNNs becomes more important, especially in safety-critical applications (e.g., medical diagnosis, autonomous driving). In this paper, we propose DeepVID, a Deep learning approach to Visually Interpret and Diagnose DNN models, especially image classifiers. In detail, we train a small locally-faithful model to mimic the behavior of an original cumbersome DNN around a particular data instance of interest, and the local model is sufficiently simple such that it can be visually interpreted (e.g., a linear model). Knowledge distillation is used to transfer the knowledge from the cumbersome DNN to the small model, and a deep generative model (i.e., variational auto-encoder) is used to generate neighbors around the instance of interest. Those neighbors, which come with small feature variances and semantic meanings, can effectively probe the DNN's behaviors around the interested instance and help the small model to learn those behaviors. Through comprehensive evaluations, as well as case studies conducted together with deep learning experts, we validate the effectiveness of DeepVID. [ABSTRACT FROM AUTHOR]

Ensemble simulations are used in climate research to account for natural variability. For medium-term decadal predictions, each simulation run is initialized with real observations from a different day resulting in a set of possible climatic futures. Understanding the variability and the predictive power in this wealth of data is still a challenging task. In this paper, we introduce a visual analytics system to explore variability within ensembles of decadal climate predictions. We propose a new interactive visualization technique (clustering timeline) based on the Sankey diagram, which conveys a concise summary of data similarity and its changes over time. We augment the system with two additional visualizations, filled contour maps and heatmaps, to provide analysts with additional information relating the new diagram to raw data and automatic clustering results. The usefulness of the technique is demonstrated by case studies and user interviews. [ABSTRACT FROM AUTHOR]

In this paper, we present a data-flow system which supports comparative analysis of time-dependent data and interactive simulation steering. The system creates data on-the-fly to allow for the exploration of different parameters and the investigation of multiple scenarios. Existing data-flow architectures provide no generic approach to handle modules that perform complex temporal processing such as particle tracing or statistical analysis over time. Moreover, there is no solution to create and manage module data, which is associated with alternative scenarios. Our solution is based on generic data-flow algorithms to automate this process, enabling elaborate data-flow procedures, such as simulation, temporal integration or data aggregation over many time steps in many worlds. To hide the complexity from the user, we extend the World Lines interaction techniques to control the novel data-flow architecture. The concept of multiple, special-purpose cursors is introduced to let users intuitively navigate through time and alternative scenarios. Users specify only what they want to see, the decision which data are required is handled automatically. The concepts are explained by taking the example of the simulation and analysis of material transport in levee-breach scenarios. To strengthen the general applicability, we demonstrate the investigation of vortices in an offline-simulated dam-break data set. [ABSTRACT FROM AUTHOR]

Multimodal sentiment analysis aims to recognize people's attitudes from multiple communication channels such as verbal content (i.e., text), voice, and facial expressions. It has become a vibrant and important research topic in natural language processing. Much research focuses on modeling the complex intra- and inter-modal interactions between different communication channels. However, current multimodal models with strong performance are often deep-learning-based techniques and work like black boxes. It is not clear how models utilize multimodal information for sentiment predictions. Despite recent advances in techniques for enhancing the explainability of machine learning models, they often target unimodal scenarios (e.g., images, sentences), and little research has been done on explaining multimodal models. In this paper, we present an interactive visual analytics system, M2 Lens, to visualize and explain multimodal models for sentiment analysis. M2 Lens provides explanations on intra- and inter-modal interactions at the global, subset, and local levels. Specifically, it summarizes the influence of three typical interaction types (i.e., dominance, complement, and conflict) on the model predictions. Moreover, M2 Lens identifies frequent and influential multimodal features and supports the multi-faceted exploration of model behaviors from language, acoustic, and visual modalities. Through two case studies and expert interviews, we demonstrate our system can help users gain deep insights into the multimodal models for sentiment analysis. [ABSTRACT FROM AUTHOR]

Hands-on training is an effective way to practice theoretical cybersecurity concepts and increase participants’ skills. In this article, we discuss the application of visual analytics principles to the design, execution, and evaluation of training sessions. We propose a conceptual model employing visual analytics that supports the sensemaking activities of users involved in various phases of the training life cycle. The model emerged from our long-term experience in designing and organizing diverse hands-on cybersecurity training sessions. It provides a classification of visualizations and can be used as a framework for developing novel visualization tools supporting phases of the training life-cycle. We demonstrate the model application on examples covering two types of cybersecurity training programs. [ABSTRACT FROM AUTHOR]

Advances in language modeling have led to the development of deep attention-based models that are performant across a wide variety of natural language processing (NLP) problems. These language models are typified by a pre-training process on large unlabeled text corpora and subsequently fine-tuned for specific tasks. Although considerable work has been devoted to understanding the attention mechanisms of pre-trained models, it is less understood how a model's attention mechanisms change when trained for a target NLP task. In this paper, we propose a visual analytics approach to understanding fine-tuning in attention-based language models. Our visualization, Attention Flows, is designed to support users in querying, tracing, and comparing attention within layers, across layers, and amongst attention heads in Transformer-based language models. To help users gain insight on how a classification decision is made, our design is centered on depicting classification-based attention at the deepest layer and how attention from prior layers flows throughout words in the input. Attention Flows supports the analysis of a single model, as well as the visual comparison between pre-trained and fine-tuned models via their similarities and differences. We use Attention Flows to study attention mechanisms in various sentence understanding tasks and highlight how attention evolves to address the nuances of solving these tasks. [ABSTRACT FROM AUTHOR]

Visual data analysis often requires grouping of data objects based on their similarity. In many application domains researchers use algorithms and techniques like clustering and multidimensional scaling to extract groupings from data. While extracting these groups using a single similarity criteria is relatively straightforward, comparing alternative criteria poses additional challenges. In this paper we define visual reconciliation as the problem of reconciling multiple alternative similarity spaces through visualization and interaction. We derive this problem from our work on model comparison in climate science where climate modelers are faced with the challenge of making sense of alternative ways to describe their models: one through the output they generate, another through the large set of properties that describe them. Ideally, they want to understand whether groups of models with similar spatio-temporal behaviors share similar sets of criteria or, conversely, whether similar criteria lead to similar behaviors. We propose a visual analytics solution based on linked views, that addresses this problem by allowing the user to dynamically create, modify and observe the interaction among groupings, thereby making the potential explanations apparent. We present case studies that demonstrate the usefulness of our technique in the area of climate science. [ABSTRACT FROM PUBLISHER]

We define behavior as a set of actions performed by some actor during a period of time. We consider the problem of analyzing a large collection of behaviors by multiple actors, more specifically, identifying typical behaviors and spotting anomalous behaviors. We propose an approach leveraging topic modeling techniques – LDA (Latent Dirichlet Allocation) Ensembles – to represent categories of typical behaviors by topics that are obtained through topic modeling a behavior collection. When such methods are applied to text in natural languages, the quality of the extracted topics are usually judged based on the semantic relatedness of the terms pertinent to the topics. This criterion, however, is not necessarily applicable to topics extracted from non-textual data, such as action sets, since relationships between actions may not be obvious. We have developed a suite of visual and interactive techniques supporting the construction of an appropriate combination of topics based on other criteria, such as distinctiveness and coverage of the behavior set. Two case studies on analyzing operation behaviors in the security management system and visiting behaviors in an amusement park, and the expert evaluation of the first case study demonstrate the effectiveness of our approach. [ABSTRACT FROM AUTHOR]

With the rapid adoption of machine learning techniques for large-scale applications in science and engineering comes the convergence of two grand challenges in visualization. First, the utilization of black box models (e.g., deep neural networks) calls for advanced techniques in exploring and interpreting model behaviors. Second, the rapid growth in computing has produced enormous datasets that require techniques that can handle millions or more samples. Although some solutions to these interpretability challenges have been proposed, they typically do not scale beyond thousands of samples, nor do they provide the high-level intuition scientists are looking for. Here, we present the first scalable solution to explore and analyze high-dimensional functions often encountered in the scientific data analysis pipeline. By combining a new streaming neighborhood graph construction, the corresponding topology computation, and a novel data aggregation scheme, namely topology aware datacubes, we enable interactive exploration of both the topological and the geometric aspect of high-dimensional data. Following two use cases from high-energy-density (HED) physics and computational biology, we demonstrate how these capabilities have led to crucial new insights in both applications. [ABSTRACT FROM AUTHOR]

Machine learning models are currently being deployed in a variety of real-world applications where model predictions are used to make decisions about healthcare, bank loans, and numerous other critical tasks. As the deployment of artificial intelligence technologies becomes ubiquitous, it is unsurprising that adversaries have begun developing methods to manipulate machine learning models to their advantage. While the visual analytics community has developed methods for opening the black box of machine learning models, little work has focused on helping the user understand their model vulnerabilities in the context of adversarial attacks. In this paper, we present a visual analytics framework for explaining and exploring model vulnerabilities to adversarial attacks. Our framework employs a multi-faceted visualization scheme designed to support the analysis of data poisoning attacks from the perspective of models, data instances, features, and local structures. We demonstrate our framework through two case studies on binary classifiers and illustrate model vulnerabilities with respect to varying attack strategies. [ABSTRACT FROM AUTHOR]

It is hard to efficiently model the light transport in scenes with translucent objects for interactive applications. The inter-reflection between objects and their environments and the subsurface scattering through the materials intertwine to produce visual effects like color bleeding, light glows, and soft shading. Monte-Carlo based approaches have demonstrated impressive results but are computationally expensive, and faster approaches model either only inter-reflection or only subsurface scattering. In this paper, we present a simple analytic model that combines diffuse inter-reflection and isotropic subsurface scattering. Our approach extends the classical work in radiosity by including a subsurface scattering matrix that operates in conjunction with the traditional form factor matrix. This subsurface scattering matrix can be constructed using analytic, measurement-based or simulation-based models and can capture both homogeneous and heterogeneous translucencies. Using a fast iterative solution to radiosity, we demonstrate scene relighting and dynamically varying object translucencies at near interactive rates. [ABSTRACT FROM AUTHOR]

Topic modeling has been widely used for analyzing text document collections. Recently, there have been significant advancements in various topic modeling techniques, particularly in the form of probabilistic graphical modeling. State-of-the-art techniques such as Latent Dirichlet Allocation (LDA) have been successfully applied in visual text analytics. However, most of the widely-used methods based on probabilistic modeling have drawbacks in terms of consistency from multiple runs and empirical convergence. Furthermore, due to the complicatedness in the formulation and the algorithm, LDA cannot easily incorporate various types of user feedback. To tackle this problem, we propose a reliable and flexible visual analytics system for topic modeling called UTOPIAN (User-driven Topic modeling based on Interactive Nonnegative Matrix Factorization). Centered around its semi-supervised formulation, UTOPIAN enables users to interact with the topic modeling method and steer the result in a user-driven manner. We demonstrate the capability of UTOPIAN via several usage scenarios with real-world document corpuses such as InfoVis/VAST paper data set and product review data sets. [ABSTRACT FROM AUTHOR]

As increasing volumes of urban data are captured and become available, new opportunities arise for data-driven analysis that can lead to improvements in the lives of citizens through evidence-based decision making and policies. In this paper, we focus on a particularly important urban data set: taxi trips. Taxis are valuable sensors and information associated with taxi trips can provide unprecedented insight into many different aspects of city life, from economic activity and human behavior to mobility patterns. But analyzing these data presents many challenges. The data are complex, containing geographical and temporal components in addition to multiple variables associated with each trip. Consequently, it is hard to specify exploratory queries and to perform comparative analyses (e.g., compare different regions over time). This problem is compounded due to the size of the data-there are on average 500,000 taxi trips each day in NYC. We propose a new model that allows users to visually query taxi trips. Besides standard analytics queries, the model supports origin-destination queries that enable the study of mobility across the city. We show that this model is able to express a wide range of spatio-temporal queries, and it is also flexible in that not only can queries be composed but also different aggregations and visual representations can be applied, allowing users to explore and compare results. We have built a scalable system that implements this model which supports interactive response times; makes use of an adaptive level-of-detail rendering strategy to generate clutter-free visualization for large results; and shows hidden details to the users in a summary through the use of overlay heat maps. We present a series of case studies motivated by traffic engineers and economists that show how our model and system enable domain experts to perform tasks that were previously unattainable for them. [ABSTRACT FROM AUTHOR]

In this paper, we introduce ParaGlide, a visualization system designed for interactive exploration of parameter spaces of multidimensional simulation models. To get the right parameter configuration, model developers frequently have to go back and forth between setting input parameters and qualitatively judging the outcomes of their model. Current state-of-the-art tools and practices, however, fail to provide a systematic way of exploring these parameter spaces, making informed decisions about parameter configurations a tedious and workload-intensive task. ParaGlide endeavors to overcome this shortcoming by guiding data generation using a region-based user interface for parameter sampling and then dividing the model's input parameter space into partitions that represent distinct output behavior. In particular, we found that parameter space partitioning can help model developers to better understand qualitative differences among possibly high-dimensional model outputs. Further, it provides information on parameter sensitivity and facilitates comparison of models. We developed ParaGlide in close collaboration with experts from three different domains, who all were involved in developing new models for their domain. We first analyzed current practices of six domain experts and derived a set of tasks and design requirements, then engaged in a user-centered design process, and finally conducted three longitudinal in-depth case studies underlining the usefulness of our approach. [ABSTRACT FROM AUTHOR]

Many approaches for analyzing a high-dimensional dataset assume that the dataset contains specific structures, e.g., clusters in linear subspaces or non-linear manifolds. This yields a trial-and-error process to verify the appropriate model and parameters. This paper contributes an exploratory interface that supports visual identification of low-dimensional structures in a high-dimensional dataset, and facilitates the optimized selection of data models and configurations. Our key idea is to abstract a set of global and local feature descriptors from the neighborhood graph-based representation of the latent low-dimensional structure, such as pairwise geodesic distance (GD) among points and pairwise local tangent space divergence (LTSD) among pointwise local tangent spaces (LTS). We propose a new LTSD-GD view, which is constructed by mapping LTSD and GD to the $x$ axis and $y$ axis using 1D multidimensional scaling, respectively. Unlike traditional dimensionality reduction methods that preserve various kinds of distances among points, the LTSD-GD view presents the distribution of pointwise LTS ($x$ axis) and the variation of LTS in structures (the combination of $x$ axis and $y$ axis). We design and implement a suite of visual tools for navigating and reasoning about intrinsic structures of a high-dimensional dataset. Three case studies verify the effectiveness of our approach. [ABSTRACT FROM PUBLISHER]

Distributions are often used to model uncertainty in many scientific datasets. To preserve the correlation among the spatially sampled grid locations in the dataset, various standard multivariate distribution models have been proposed in visualization literature. These models treat each grid location as a univariate random variable which models the uncertainty at that location. Standard multivariate distributions (both parametric and nonparametric) assume that all the univariate marginals are of the same type/family of distribution. But in reality, different grid locations show different statistical behavior which may not be modeled best by the same type of distribution. In this paper, we propose a new multivariate uncertainty modeling strategy to address the needs of uncertainty modeling in scientific datasets. Our proposed method is based on a statistically sound multivariate technique called Copula, which makes it possible to separate the process of estimating the univariate marginals and the process of modeling dependency, unlike the standard multivariate distributions. The modeling flexibility offered by our proposed method makes it possible to design distribution fields which can have different types of distribution (Gaussian, Histogram, KDE etc.) at the grid locations, while maintaining the correlation structure at the same time. Depending on the results of various standard statistical tests, we can choose an optimal distribution representation at each location, resulting in a more cost efficient modeling without significantly sacrificing on the analysis quality. To demonstrate the efficacy of our proposed modeling strategy, we extract and visualize uncertain features like isocontours and vortices in various real world datasets. We also study various modeling criterion to help users in the task of univariate model selection. [ABSTRACT FROM PUBLISHER]

Recently proposed techniques have finally made it possible for analysts to interactively explore very large datasets in real time. However powerful, the class of analyses these systems enable is somewhat limited: specifically, one can only quickly obtain plots such as histograms and heatmaps. In this paper, we contribute Gaussian Cubes, which significantly improves on state-of-the-art systems by providing interactive modeling capabilities, which include but are not limited to linear least squares and principal components analysis (PCA). The fundamental insight in Gaussian Cubes is that instead of precomputing counts of many data subsets (as state-of-the-art systems do), Gaussian Cubes precomputes the best multivariate Gaussian for the respective data subsets. As an example, Gaussian Cubes can fit hundreds of models over millions of data points in well under a second, enabling novel types of visual exploration of such large datasets. We present three case studies that highlight the visualization and analysis capabilities in Gaussian Cubes, using earthquake safety simulations, astronomical catalogs, and transportation statistics. The dataset sizes range around one hundred million elements and 5 to 10 dimensions. We present extensive performance results, a discussion of the limitations in Gaussian Cubes, and future research directions. [ABSTRACT FROM AUTHOR]

Visual analytics techniques help users explore high-dimensional data. However, it is often challenging for users to express their domain knowledge in order to steer the underlying data model, especially when they have little attribute-level knowledge. Furthermore, users' complex, high-level domain knowledge, compared to low-level attributes, posits even greater challenges. To overcome these challenges, we introduce a technique to interpret a user's drawings with an interactive, nonlinear axis mapping approach called AxiSketcher. This technique enables users to impose their domain knowledge on a visualization by allowing interaction with data entries rather than with data attributes. The proposed interaction is performed through directly sketching lines over the visualization. Using this technique, users can draw lines over selected data points, and the system forms the axes that represent a nonlinear, weighted combination of multidimensional attributes. In this paper, we describe our techniques in three areas: 1) the design space of sketching methods for eliciting users' nonlinear domain knowledge; 2) the underlying model that translates users' input, extracts patterns behind the selected data points, and results in nonlinear axes reflecting users' complex intent; and 3) the interactive visualization for viewing, assessing, and reconstructing the newly formed, nonlinear axes. [ABSTRACT FROM AUTHOR]

Despite the recent popularity of visual analytics focusing on big data, little is known about how to support users that use visualization techniques to explore multi-dimensional datasets and accomplish specific tasks. Our lack of models that can assist end-users during the data exploration process has made it challenging to learn from the user's interactive and analytical process. The ability to model how a user interacts with a specific visualization technique and what difficulties they face are paramount in supporting individuals with discovering new patterns within their complex datasets. This paper introduces the notion of visualization systems understanding and modeling user interactions with the intent of guiding a user through a task thereby enhancing visual data exploration. The challenges faced and the necessary future steps to take are discussed; and to provide a working example, a grammar-based model is presented that can learn from user interactions, determine the common patterns among a number of subjects using a K-Reversible algorithm, build a set of rules, and apply those rules in the form of suggestions to new users with the goal of guiding them along their visual analytic process. A formal evaluation study with 300 subjects was performed showing that our grammar-based model is effective at capturing the interactive process followed by users and that further research in this area has the potential to positively impact how users interact with a visualization system. [ABSTRACT FROM AUTHOR]

In this article we demonstrate robust estimation of the model parameters of a fully-linear data-driven BRDF model from a reflectance map under known natural lighting. To regularize the estimation of the model parameters, we leverage the reflectance similarities within a material class. We approximate the space of homogeneous BRDFs using a Gaussian mixture model, and assign a material class to each Gaussian in the mixture model. We formulate the estimation of the model parameters as a non-linear maximum a-posteriori optimization, and introduce a linear approximation that estimates a solution per material class from which the best solution is selected. We demonstrate the efficacy and robustness of our method using the MERL BRDF database under a variety of natural lighting conditions, and we provide a proof-of-concept real-world experiment. [ABSTRACT FROM AUTHOR]

Visualization and analysis techniques play a key role in the discovery of relevant features in ensemble data. Trends, in the form of persisting commonalities or differences in time-varying ensemble datasets, constitute one of the most expressive feature types in ensemble analysis. We develop a flow-graph representation as the core of a system designed for the visual analysis of trends in time-varying ensembles. In our interactive analysis framework, this graph is linked to a representation of ensemble parameter-space and the ensemble itself. This facilitates a detailed examination of trends and their correlations to properties of input-space. We demonstrate the utility of the proposed trends analysis framework in several benchmark data sets, highlighting its capability to support goal-driven design of time-varying simulations. [ABSTRACT FROM AUTHOR]

Topic modeling, a method of statistically extracting thematic content from a large collection of texts, is used for a wide variety of tasks within text analysis. Though there are a growing number of tools and techniques for exploring single models, comparisons between models are generally reduced to a small set of numerical metrics. These metrics may or may not reflect a model's performance on the analyst's intended task, and can therefore be insufficient to diagnose what causes differences between models. In this paper, we explore task-centric topic model comparison, considering how we can both provide detail for a more nuanced understanding of differences and address the wealth of tasks for which topic models are used. We derive comparison tasks from single-model uses of topic models, which predominantly fall into the categories of understanding topics, understanding similarity, and understanding change. Finally, we provide several visualization techniques that facilitate these tasks, including buddy plots, which combine color and position encodings to allow analysts to readily view changes in document similarity. [ABSTRACT FROM AUTHOR]

Visual analytics supports humans in generating knowledge from large and often complex datasets. Evidence is collected, collated and cross-linked with our existing knowledge. In the process, a myriad of analytical and visualisation techniques are employed to generate a visual representation of the data. These often introduce their own uncertainties, in addition to the ones inherent in the data, and these propagated and compounded uncertainties can result in impaired decision making. The user's confidence or trust in the results depends on the extent of user's awareness of the underlying uncertainties generated on the system side. This paper unpacks the uncertainties that propagate through visual analytics systems, illustrates how human's perceptual and cognitive biases influence the user's awareness of such uncertainties, and how this affects the user's trust building. The knowledge generation model for visual analytics is used to provide a terminology and framework to discuss the consequences of these aspects in knowledge construction and though examples, machine uncertainty is compared to human trust measures with provenance. Furthermore, guidelines for the design of uncertainty-aware systems are presented that can aid the user in better decision making. [ABSTRACT FROM AUTHOR]

Online news, microblogs and other media documents all contain valuable insight regarding events and responses to events. Underlying these documents is the concept of framing, a process in which communicators act (consciously or unconsciously) to construct a point of view that encourages facts to be interpreted by others in a particular manner. As media discourse evolves, how topics and documents are framed can undergo change, shifting the discussion to different viewpoints or rhetoric. What causes these shifts can be difficult to determine directly; however, by linking secondary datasets and enabling visual exploration, we can enhance the hypothesis generation process. In this paper, we present a visual analytics framework for event cueing using media data. As discourse develops over time, our framework applies a time series intervention model which tests to see if the level of framing is different before or after a given date. If the model indicates that the times before and after are statistically significantly different, this cues an analyst to explore related datasets to help enhance their understanding of what (if any) events may have triggered these changes in discourse. Our framework consists of entity extraction and sentiment analysis as lenses for data exploration and uses two different models for intervention analysis. To demonstrate the usage of our framework, we present a case study on exploring potential relationships between climate change framing and conflicts in Africa. [ABSTRACT FROM AUTHOR]