Search

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.

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., Comment: This paper was submitted to IEEE VIS 2021, and it was revised after receiving the first round reviews. The authors would like to thank the anonymous reviewers for their effort, comments, and suggestions. The work is part of the big effort by many RAMPVIS colleagues [arXiv:2012.04757]

A graph is a mathematical model for representing a system of pairwise relationships between entities. The term “graph” or “graph data” is quite often used to refer, actually, to a system of relationships, which can be represented as a graph, rather than to the mathematical model itself. In line with this practice, the term “graph” is used in this chapter as a synonym to “system of relationships”. Graph data have high importance in many application domains, such as social network analysis, transport network analysis, systems biology, transaction analysis, to name a few. The relational nature of graph data sets presents unique challenges that open up space for distinctive and innovative visual analytics methods. Graphs often contain inherent hierarchies, and the analysis outcomes vary significantly depending on the scale (i.e., aggregation level) at which the relationships are considered. Identification, analysis, and interpretation of inherent structures, such as tightly connected parts of the network, requires a combination of algorithmic and interactive methods. Analysis of graph structure may require involvement of different graph theoretic properties of nodes and links, consideration of paths, and extraction of recurring connectivity patterns. All of these analyses are taken to a different level of complexity when these graphs are multivariate, i.e., where each node carries multiple attributes, and dynamic, i.e., when the structure of the graph changes with time. In this chapter, we discuss these challenges in depth and present examples of employing visual analytics techniques for extracting valuable findings from this highly rich and interesting data type.

Texts are created for humans, who are trained to read and understand them. Texts are poorly suited for machine processing; still, humans need computer help when it is necessary to gain an overall understanding of characteristics and contents of large volumes of text or to find specific information in these volumes. Computer support in text analysis involves derivation of various kinds of structured data, such as numeric attributes and lists of significant items with associated numeric measures or weighted binary relationships between them. Computers themselves cannot give any meaning to the data they derive; therefore, the data need to be presented to humans in ways enabling semantic interpretations. While there exist a few text-specific visualisation techniques, such as Word Cloud and Word Tree, which explicitly represent words, it is often beneficial to use also general approaches suitable for multidimensional data. Collections of texts having spatial and/or temporal references are transformed to data that can be visualised and analysed using general methods devised for spatial, temporal, and spatio-temporal data.We show multiple examples of possible tasks in text data analysis and approaches to accomplishing them.

Images and video recordings are commonly categorised as unstructured data, which means that they are not primarily suited for computer analysis. The contents of unstructured data cannot be adequately represented by numbers or symbols and require the power of human vision for extracting meaningful information. While images and video are well suited for human visual perception, coping with large amounts of unstructured data may be daunting for humans. Hence, humans need computer support but cannot be substituted by computers.

Visual analytics approaches combine interactive visualisations with the use of computational techniques for data processing and analysis. Combining visualisation and computation has two sides. One side is computational support to visual analysis: outcomes of computations are intended to provide input to human cognition; for this purpose, they are represented visually. The other side is visual support to application of computational methods, which includes visual exploration of data properties for preparing data to computations, evaluation of computation outcomes, and comparison of results of different runs of computational techniques. This chapter focuses in more detail on the computational support to visual analysis. A major common purpose for using computational methods in visualisation is enabling an overview of voluminous and complex data. The general approaches are spatialisation, which is achieved by means of data embedding techniques, and grouping, which is achieved using clustering algorithms. Distance functions are auxiliary computational techniques used both for data embedding and clustering. They provide numeric assessments of the dissimilarity between data items. Another class of auxiliary techniques is feature selection. The chapter describes the use of data embedding and clustering. By example of clustering, the general principles of visual analytics support to application of computational methods are demonstrated. Topic modelling is a special group of data embedding techniques originally designed for textual data but applicable also to other kinds of data. The main ideas, properties, and uses of the topic modelling methods are discussed in a separate section.

There are two major types of temporal data, events and time series of attribute values, and there are methods for transforming one of them into the other. For events, a general analysis task is to understand how they are distributed in time. For time series, as well as for events of diverse kinds, a general task is to understand how the attribute values or the kinds of occurring events vary over time. In analysing temporal distributions and variations, it is essential to account for the specific features of time and temporal phenomena, particularly, recurring cycles and temporal dependence. To see patterns of temporal distribution or variation, people commonly apply visual displays where one dimension represents time and the other is used to show individual events, attribute values, or statistical summaries. To see the data in the context of temporal cycles, a common approach is to use a 2D display where one or two cycles are represented by display dimensions. For large and/or complex data, visual displays need to be combined with techniques for computational analysis, such as clustering, embedding, sequence mining, and motif discovery. We show and discuss examples of employing such combinations in application to different data and analysis tasks.

In this chapter, we discuss how visual analytics techniques can support you in investigating and understanding the properties of your data and in conducting common data processing tasks. We consider several examples of possible problems in data and how they may manifest in visual representations, discuss where and why data quality issues can appear, and introduce a number of elementary data processing operations and computational techniques that help, in combination with visualisation, understand data characteristics and detect abnormalities.

We begin with a simple motivating example that shows how putting spatial data on a map and seeing spatial relationships can help an analyst to make important discoveries. We consider possible contents and forms of spatial data, the ways of specifying spatial locations, and how to use spatial references for joining different datasets. We discuss the specifics of the (geographic) space and spatial phenomena, where spatial relationships within and between the phenomena play a crucial role. The First Law of Geography states: “Everything is related to everything else, but near things are more related than distant things”, emphasising the importance of distance relationships. However, the spatial context, which includes the properties of the underlying space and the things and phenomena existing around, often modifies these relationships; hence, everything related to space needs to be considered in its spatial context. We describe techniques for transforming and analysing spatial data and give an example of an analytical workflow where some of these techniques are used but, as usual, the main instruments of the analysis are human background knowledge, capability to see and interpret patterns, and reasoning.

Data scientists usually aim at building computer models. Computeroriented modelling methods and software tools are developed in statistics, machine learning, data mining, and various specialised disciplines, such as spatial statistics, transportation research, and animal ecology. However, valid and useful computerbased models cannot be obtained by mere application of some modelling software to available data. Modelling requires understanding of the phenomenon that is modelled, the available data, and the model produced by the computer, which means that computer modelling is a process requiring involvement of a human. This chapter formulates the principles of thoughtful model building and includes examples of the use of visual analytics approaches for fulfilling these principles. The examples cover the tasks of feature engineering and selection, iterative model evaluation and progressive improvement, and comparison of different models.

One very common challenge that every data scientists has to deal with is to make sense of data sets with many attributes, where “many” can sometimes be tens, sometimes hundreds, and even thousands. Whether your goal is to do exploratory analysis on the relationships between the attributes, or to build models of the underlying phenomena, working with many dimensions is not trivial. The high number of attributes is a barrier against using some of the standard visual representations: just try to imagine a scatterplot matrix where you want to look at the pairwise distributions combinations of 100 variables. Moreover, any computational method that you apply produces results that are challenging to interpret. Even linear regression, one of the easiest models to understand, becomes quite complex if you need to investigate the interactions between hundreds of variables. This chapter will discuss how not to get lost in these high-dimensional spaces and how visual analytics techniques can help you navigate your way through.

We introduce the basic principles and rules of the visual representation of information. Any visualisation involves so-called visual variables, such as position along an axis, size, colour hue and lightness, and shape of a graphical element. The variables differ by their perceptual properties, and it is important to choose appropriate variables depending on the properties of the data they are intended to represent. We discuss the commonly used types of plots, graphs, charts, and other displays telling what kinds of information each of them is suited for and what visual variables it involves. We explain the necessity of supporting the use of visual representations by tools for interacting with the displays and the information they represent and introduce the common interaction techniques. We also discuss the limitations of visualisation and interaction, substantiating the need to combine them with computational operations for data processing and analysis.

An illustrated example of problem solving is meant to demonstrate how visual representations of data support human reasoning and deriving knowledge from data.We argue that human reasoning plays a crucial role in solving non-trivial problems. Even when the primary goal of data analysis is to create a predictive model to be executed by computers, this cannot be done without human reasoning and derivation of new knowledge, which includes understanding of the analysis subject and knowledge of the computer model built. Reasoning requires conveying information to the human’s mind, and visual representations are best suited for this. Visual analytics focuses on supporting human analytical reasoning and develops approaches combining visualisations, interactive operations, and computational processing. The underlying idea is to enable synergistic joint work of humans and computers, in which each side can effectively utilise its unique capabilities. The ideas and approaches of visual analytics are therefore very relevant to data science.

There are different kinds of spatio-temporal phenomena, including events that occur at different locations, movements of discrete entities, changes of shapes and sizes of entities, changes of conditions at different places and overall situations across large areas. Spatio-temporal data may specify positions, times, and characteristics of spatial events, represent trajectories of moving entities, or consist of spatially referenced time series of attribute values. It is possible to transform spatiotemporal data from one of the forms to another and thus adapt them to analysis tasks. After presenting a motivating example of analysis, we discuss the specifics of spatio-temporal data and possible data quality issues that often appear in real data sets and influence analysis processes and results.We introduce and discuss the visual analytics techniques suitable for spatio-temporal data and present another example of an analytical workflow.

Visual analytics systems combine machine learning or other analytic techniques with interactive data visualization to promote sensemaking and analytical reasoning. It is through such techniques that people can make sense of large, complex data. While progress has been made, the tactful combination of machine learning and data visualization is still under-explored. This state-of-the-art report presents a summary of the progress that has been made by highlighting and synthesizing select research advances. Further, it presents opportunities and challenges to enhance the synergy between machine learning and visual analytics for impactful future research directions.

Fundamental to the effective use of visualization as an analytic and descriptive tool is the assurance that presenting data visually provides the capability of making inferences from what we see. This paper explores two related approaches to quantifying the confidence we may have in making visual inferences from mapped geospatial data. We adapt Wickham et al. 's ‘Visual Line-up’ method as a direct analogy with Null Hypothesis Significance Testing (NHST) and propose a new approach for generating more credible spatial null hypotheses. Rather than using as a spatial null hypothesis the unrealistic assumption of complete spatial randomness, we propose spatially autocorrelated simulations as alternative nulls. We conduct a set of crowdsourced experiments (n=361) to determine the just noticeable difference (JND) between pairs of choropleth maps of geographic units controlling for spatial autocorrelation (Moran's I statistic) and geometric configuration (variance in spatial unit area). Results indicate that people's abilities to perceive differences in spatial autocorrelation vary with baseline autocorrelation structure and the geometric configuration of geographic units. These results allow us, for the first time, to construct a visual equivalent of statistical power for geospatial data. Our JND results add to those provided in recent years by Klippel et al. (2011), Harrison et al. (2014) and Kay & Heer (2015) for correlation visualization. Importantly, they provide an empirical basis for an improved construction of visual line-ups for maps and the development of theory to inform geospatial tests of graphical inference.

Action sequences, where atomic user actions are represented in a labelled, timestamped form, are becoming a fundamental data asset in the inspection and monitoring of user behaviour in digital systems. Although the analysis of such sequences is highly critical to the investigation of activities in cyber security applications, existing solutions fail to provide a comprehensive understanding due to the complex semantic and temporal characteristics of these data. This paper presents a visual analytics approach that aims to facilitate a user-involved, multi-faceted decision making process during the identification and the investigation of “unusual” action sequences. We first report the results of the task analysis and domain characterisation process. Then we describe the components of our multi-level analysis approach that comprises of constraint-based sequential pattern mining and semantic distance based clustering, and multi-scalar visualisations of users and their sequences. Finally, we demonstrate the applicability of our approach through a case study that involves tasks requiring effective decision-making by a group of domain experts. Although our solution here is tightly informed by a user-centred, domain-focused design process, we present findings and techniques that are transferable to other applications where the analysis of such sequences is of interest.

Promoting a wider range of contribution types can facilitate healthy growth of the visualization community, while increasing the intellectual diversity of visualization research papers. In this article, we discuss the importance of contribution types and summarize contribution types that can be meaningful in visualization research. We also propose several concrete next steps we can and should take to ensure a successful launch of the contribution types.

User behavior analysis is complex and especially crucial in the cyber security domain. Understanding dynamic and multi-variate user behavior are challenging. Traditional sequential and timeline based method cannot easily address the complexity of temporal and relational features of user behaviors. We propose a map-based visual metaphor and create an interactive map for encoding user behaviors. It enables analysts to explore and identify user behavior patterns and helps them to understand why some behaviors are regarded as anomalous. We experiment with a real dataset containing multiple user sessions, consisting of sequences of diverse types of actions. In the behavior map, we encode an action as a city and user sessions as trajectories going through the cities. The position of the cities is determined by the sequential and temporal relationship of actions. Spatial and temporal patterns on the map reflect behavior patterns in the action space. In the case study, we illustrate how we explore relationships between actions, identify patterns of the typical session and detect anomaly behaviors.

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.

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. Making decisions based on such systems; however, requires an in-depth understanding of user behaviour both at an individual and at a group level where a group can consist of users with similar roles. We present a visual analytics approach to help analysts gain a comprehensive, multifaceted understanding of user behaviour at multiple levels. 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, where the profiles are built based on features derived from sessions they perform and visualised with task-informed designs to facilitate interactive exploration and investigation. We also present techniques to extract user tasks that summarise the behaviour and to cluster users according to these tasks for providing hierarchical overviews of groups of users along with individual users and the sessions they conduct. We externalise a series of analysis goals and tasks, and evaluate our methods through a number of use cases that demonstrate how these tasks are addressed. We observe that with the aid of interactive visual hierarchical user profiles, analysts were 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.

Comparing multiple variables to select those that effectively characterize complex entities is important in a wide variety of domains – geodemographics for example. Identifying variables that correlate is a common practice to remove redundancy, but correlation varies across space, with scale and over time, and the frequently used global statistics hide potentially important differentiating local variation. For more comprehensive and robust insights into multivariate relations, these local correlations need to be assessed through various means of defining locality. We explore the geography of this issue, and use novel interactive visualization to identify interdependencies in multivariate data sets to support geographically informed multivariate analysis. We offer terminology for considering scale and locality, visual techniques for establishing the effects of scale on correlation and a theoretical framework through which variation in geographic correlation with scale and locality are addressed explicitly. Prototype software demonstrates how these contributions act together. These techniques enable multiple variables and their geographic characteristics to be considered concurrently as we extend visual parameter space analysis (vPSA) to the spatial domain. We find variable correlations to be sensitive to scale and geography to varying degrees in the context of energy-based geodemographics. This sensitivity depends upon the calculation of locality as well as the geographical and statistical structure of the variable.

The analysis of financial assets' correlations is fundamental to many aspects of finance theory and practice, especially modern portfolio theory and the study of risk. In order to manage investment risk, in-depth analysis of changing correlations is needed, with both high and low correlations between financial assets (and groups thereof) important to identify. In this paper, we propose a visual analytics framework for the interactive analysis of relations and structures in dynamic, high-dimensional correlation data. We conduct a series of interviews and review the financial correlation analysis literature to guide our design. Our solution combines concepts from multi-dimensional scaling, weighted complete graphs and threshold networks to present interactive, animated displays which use proximity as a visual metaphor for correlation and animation stability to encode correlation stability. We devise interaction techniques coupled with context-sensitive auxiliary views to support the analysis of subsets of correlation networks. As part of our contribution, we also present behaviour profiles to help guide future users of our approach. We evaluate our approach by checking the validity of the layouts produced, presenting a number of analysis stories, and through a user study. We observe that our solutions help unravel complex behaviours and resonate well with study participants in addressing their needs in the context of correlation analysis in finance.

Across the globe, rapid growth and urbanization are placing increasing pressure on cities and governances to make the most efficient use of their resources. Some estimates predict that 70 percent of the worlds population will live in a city or suburb by 2050. One way to address this challenge is to integrate digital technology into a citys resources, assets, and infrastructure. Community services and assets that could benefit from such innovation include local governance departments, information systems, educational institutions, libraries, transportation systems, hospitals, energy suppliers, water supply networks, waste management, and law enforcement.

Visual analytics usually deals with complex data and uses sophisticated algorithmic, visual, and interactive techniques supporting the analysis. Findings and results of the analysis often need to be communicated to an audience that lacks visual analytics expertise. This requires analysis outcomes to be presented in simpler ways than that are typically used in visual analytics systems. However, not only analytical visualizations may be too complex for target audiences but also the information that needs to be presented. Analysis results may consist of multiple components, which may involve multiple heterogeneous facets. Hence, there exists a gap on the path from obtaining analysis findings to communicating them, within which two main challenges lie: information complexity and display complexity. We address this problem by proposing a general framework where data analysis and result presentation are linked by story synthesis, in which the analyst creates and organises story contents. Unlike previous research, where analytic findings are represented by stored display states, we treat findings as data constructs. We focus on selecting, assembling and organizing findings for further presentation rather than on tracking analysis history and enabling dual (i.e., explorative and communicative) use of data displays. In story synthesis, findings are selected, assembled, and arranged in meaningful layouts that take into account the structure of information and inherent properties of its components. We propose a workflow for applying the proposed conceptual framework in designing visual analytics systems and demonstrate the generality of the approach by applying it to two diverse domains, social media and movement analysis.

One of the fundamental organisational questions is how organisations identify anomalies, monitor and compare email communications between staff-staff or staff-clients or staff-customers relationships on a daily basis. The tenacious and substantial relationships are built by the combination of timely replies, frequent engagement and deep interaction between the individuals. To watchdog this periodically, we need an interactive visualisation tool that can help organisational analysts to reconnect some lost relationships and/or strengthen an existing relationship or in some cases identify inside persons (anomalies). From our point of view, Social Intelligence (SI) in an organisation is a combination of self-, social- and organisational-awareness that will help in managing complex socio-organisational changes and can be interpreted in terms of socio-organisational communication efficacy (that is, one's confidence in one's ability to deal with social and organisational information). We considered a case study, an Enron Organisation Email Scandal, to understand the relationships of staff during various parts of the years and we conducted a workshop study with legal experts to gain insights on how they carry out investigation/analysis with respect to email relationships. The outcomes of the workshop helped us develop a novel small multiples matrix-based visualisation in collaboration with our industrial partner, Red Sift UK, to find anomalies, monitor and compare how email relationships change over time and how it defines the meaning of socio-organisational communication efficacy.

The primary purpose for which statistical models are employed in the social sciences is to understand and explain phenomena occurring in the world around us. In order to be scientifically valid and actionable, the construction of such models need to be strongly informed by theory. To accomplish this, there is a need for methodologies that can enable scientists to utilise their domain knowledge effectively even in the absence of strong a priori hypotheses or whilst dealing with complex datasets containing hundreds of variables and leading to large numbers of potential models. In this paper, we describe enhanced model building processes in which we use interactive visualisations as the underlying mechanism to facilitate the construction and documentation of theory-driven models. We report our observations from a collaborative project involving social and computer scientists, and identify key roles for visualisation to support model building within the context of social science. We describe a suite of techniques to facilitate the exploration of statistical summaries of input variables, to compare the quality of alternative models, and to keep track of the model-building process. We demonstrate how these techniques operate in coordination to allow social scientists to efficiently generate models that are tightly underpinned by domain specific theory.

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.

Part 4: MAKE VIS; International audience; We describe a selection of challenges at the intersection of machine learning and data visualization and outline a subjective research agenda based on professional and personal experience. The unprecedented increase in the amount, variety and the value of data has been significantly transforming the way that scientific research is carried out and businesses operate. Within data science, which has emerged as a practice to enable this data-intensive innovation by gathering together and advancing the knowledge from fields such as statistics, machine learning, knowledge extraction, data management, and visualization, visualization plays a unique and maybe the ultimate role as an approach to facilitate the human and computer cooperation, and to particularly enable the analysis of diverse and heterogeneous data using complex computational methods where algorithmic results are challenging to interpret and operationalize. Whilst algorithm development is surely at the center of the whole pipeline in disciplines such as Machine Learning and Knowledge Discovery, it is visualization which ultimately makes the results accessible to the end user. Visualization thus can be seen as a mapping from arbitrarily high-dimensional abstract spaces to the lower dimensions and plays a central and critical role in interacting with machine learning algorithms, and particularly in interactive machine learning (iML) with including the human-in-the-loop. The central goal of the CD-MAKE VIS workshop is to spark discussions at this intersection of visualization, machine learning and knowledge discovery and bring together experts from these disciplines. This paper discusses a perspective on the challenges and opportunities in this integration of these discipline and presents a number of directions and strategies for further research.

Cohort studies in medicine are conducted to enable the study of medical hypotheses in large samples. Often, a large amount of heterogeneous data is acquired from many subjects. The analysis is usually hypothesis-driven, i.e., a specific subset of such data is studied to confirm or reject specific hypotheses. In this paper, we demonstrate how we enable the interactive visual exploration and analysis of such data, helping with the generation of new hypotheses and contributing to the process of validating them. We propose a data-cube based model which handles partially overlapping data subsets during the interactive visualization. This model enables seamless integration of the heterogeneous data, as well as linking spatial and non-spatial views on these data. We implemented this model in an application prototype, and used it to analyze data acquired in the context of a cohort study on cognitive aging. We present case-study analyses of selected aspects of brain connectivity by using the prototype implementation of the presented model, to demonstrate its potential and flexibility.

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.

Many datasets have multiple perspectives – for example space, time and description – and often analysts are required to study these multiple perspectives concurrently. This concurrent analysis becomes difficult when data are grouped and split into small multiples for comparison. A design challenge is thus to provide representations that enable multiple perspectives, split into small multiples, to be viewed simultaneously in ways that neither clutter nor overload. We present a design framework that allows us to do this. We claim that multi-perspective comparison across small multiples may be possible by superimposing perspectives on one another rather than juxtaposing those perspectives side-by-side. This approach defies conventional wisdom and likely results in visual and informational clutter. For this reason we propose designs at three levels of abstraction for each perspective. By flexibly varying the abstraction level, certain perspectives can be brought into, or out of, focus. We evaluate our framework through laboratory-style user tests. We find that superimposing, rather than juxtaposing, perspective views has little effect on performance of a low-level comparison task. We reflect on the user study and its design to further identify analysis situations for which our framework may be desirable. Although the user study findings were insufficiently discriminating, we believe our framework opens up a new design space for multi-perspective visual analysis.

Small multiples enable comparison by providing different views of a single data set in a dense and aligned manner. A common frame defines each view, which varies based upon values of a conditioning variable. An increasingly popular use of this technique is to project two-dimensional locations into a gridded space (e.g. grid maps), using the underlying distribution both as the conditioning variable and to determine the grid layout. Using whitespace in this layout has the potential to carry information, especially in a geographic context. Yet, the effects of doing so on the spatial properties of the original units are not understood. We explore the design space offered by such small multiples with gaps . We do so by constructing a comprehensive suite of metrics that capture properties of the layout used to arrange the small multiples for comparison (e.g. compactness and alignment) and the preservation of the original data (e.g. distance, topology and shape). We study these metrics in geographic data sets with varying properties and numbers of gaps. We use simulated annealing to optimize for each metric and measure the effects on the others. To explore these effects systematically, we take a new approach, developing a system to visualize this design space using a set of interactive matrices. We find that adding small amounts of whitespace to small multiple arrays improves some of the characteristics of 2D layouts, such as shape, distance and direction. This comes at the cost of other metrics, such as the retention of topology. Effects vary according to the input maps, with degree of variation in size of input regions found to be a factor. Optima exist for particular metrics in many cases, but at different amounts of whitespace for different maps. We suggest multiple metrics be used in optimized layouts, finding topology to be a primary factor in existing manually-crafted solutions, followed by a trade-off between shape and displacement. But the rich range of possible optimized layouts leads us to challenge single-solution thinking; we suggest to consider alternative optimized layouts for small multiples with gaps. Key to our work is the systematic, quantified and visual approach to exploring design spaces when facing a trade-off between many competing criteria—an approach likely to be of value to the analysis of other design spaces.

Crowds must be simulated believable in terms of their appearance and behavior to improve a virtual environment's realism. Due to the complex nature of human behavior, realistic behavior of agents in crowd simulations is still a challenging problem. In this paper, we propose a novel behavioral model which builds analytical maps to control agents’ behavior adaptively with agent–crowd interaction formulations. We introduce information theoretical concepts to construct analytical maps automatically. Our model can be integrated into crowd simulators and enhance their behavioral complexity. We made comparative analyses of the presented behavior model with measured crowd data and two agent-based crowd simulators.

Navigation and monitoring of large and crowded virtual environments is a challenging task and requires intuitive camera control techniques to assist users. In this paper, we present a novel automatic camera control technique providing a scene analysis framework based on information theory. The developed framework contains a probabilistic model of the scene to build entropy and expectancy maps. These maps are utilized to find interest points which represent either characteristic behaviors of the crowd or novel events occurring in the scene. After an interest point is chosen, the camera is updated accordingly to display this point. We tested our model in a crowd simulation environment and it performed successfully. Our method can be integrated into existent camera control modules in computer games, crowd simulations and movie pre-visualization applications.

Datasets with a large number of dimensions per data item (hundreds or more) are challenging both for computational and visual analysis. Moreover, these dimensions have different characteristics and relations that result in sub-groups and/or hierarchies over the set of dimensions. Such structures lead to heterogeneity within the dimensions. Although the consideration of these structures is crucial for the analysis, most of the available analysis methods discard the heterogeneous relations among the dimensions. In this paper, we introduce the construction and utilization of representative factors for the interactive visual analysis of structures in high-dimensional datasets. First, we present a selection of methods to investigate the sub-groups in the dimension set and associate representative factors with those groups of dimensions. Second, we introduce how these factors are included in the interactive visual analysis cycle together with the original dimensions. We then provide the steps of an analytical procedure that iteratively analyzes the datasets through the use of representative factors. We discuss how our methods improve the reliability and interpretability of the analysis process by enabling more informed selections of computational tools. Finally, we demonstrate our techniques on the analysis of brain imaging study results that are performed over a large group of subjects.

Flow data is often visualized by animated particles inserted into a flow field. The velocity of a particle on the screen is typically linearly scaled by the velocities in the data. However, the perception of velocity magnitude in animated particles is not necessarily linear. We present a study on how different parameters affect relative motion perception. We have investigated the impact of four parameters. The parameters consist of speed multiplier, direction, contrast type and the global velocity scale. In addition, we investigated if multiple motion cues, and point distribution, affect the speed estimation. Several studies were executed to investigate the impact of each parameter. In the initial results, we noticed trends in scale and multiplier. Using the trends for the significant parameters, we designed a compensation model, which adjusts the particle speed to compensate for the effect of the parameters. We then performed a second study to investigate the performance of the compensation model. From the second study we detected a constant estimation error, which we adjusted for in the last study. In addition, we connect our work to established theories in psychophysics by comparing our model to a model based on Stevens' Power Law. publishedVersion

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.

With the advance of new data acquisition and generation technologies, the biomedical domain is becoming increasingly data-driven. Thus, understanding the information in large and complex data sets has been in the focus of several research fields such as statistics, data mining, machine learning, and visualization. While the first three fields predominantly rely on computational power, visualization relies mainly on human perceptual and cognitive capabilities for extracting information. Data visualization, similar to Human–Computer Interaction, attempts an appropriate interaction between human and data to interactively exploit data sets. Specifically within the analysis of complex data sets, visualization researchers have integrated computational methods to enhance the interactive processes. In this state-of-the-art report, we investigate how such an integration is carried out. We study the related literature with respect to the underlying analytical tasks and methods of integration. In addition, we focus on how such methods are applied to the biomedical domain and present a concise overview within our taxonomy. Finally, we discuss some open problems and future challenges.

Molecular surfaces provide a useful mean for analyzing interactions between biomolecules; such as identification and characterization of ligand binding sites to a host macromolecule. We present a novel technique, which extracts potential binding sites, represented by cavities, and characterize them by 3D graphs and by amino acids. The binding sites are extracted using an implicit function sampling and graph algorithms. We propose an advanced cavity exploration technique based on the graph parameters and associated amino acids. Additionally, we interactively visualize the graphs in the context of the molecular surface. We apply our method to the analysis of MD simulations of Proteinase 3, where we verify the previously described cavities and suggest a new potential cavity to be studied. Peer Reviewed publishedVersion

High dimensional, heterogeneous datasets are challenging for domain experts to analyze. A very large number of dimensions often pose problems when visual and computational analysis tools are considered. Analysts tend to limit their attention to subsets of the data and lose potential insight in relation to the rest of the data. Generating new hypotheses is becoming problematic due to these limitations. In this paper, we discuss how interactive analysis methods can help analysts to cope with these challenges and aid them in building new hypotheses. Here, we report on the details of an analysis of data recorded in a comprehensive study of cognitive aging. We performed the analysis as a team of visualization researchers and domain experts. We discuss a number of lessons learned related to the usefulness of interactive methods in generating hypotheses.

The process of surface perception is complex and based on several influencing factors, e.g., shading, silhouettes, occluding contours, and top down cognition. The accuracy of surface perception can be measured and the influencing factors can be modified in order to decrease the error in perception. This paper presents a novel concept of how a perceptual evaluation of a visualization technique can contribute to its redesign with the aim of improving the match between the distal and the proximal stimulus. During analysis of data from previous perceptual studies, we observed that the slant of 3D surfaces visualized on 2D screens is systematically underestimated. The visible trends in the error allowed us to create a statistical model of the perceived surface slant. Based on this statistical model we obtained from user experiments, we derived a new shading model that uses adjusted surface normals and aims to reduce the error in slant perception. The result is a shape-enhancement of visualization which is driven by an experimentally-founded statistical model. To assess the efficiency of the statistical shading model, we repeated the evaluation experiment and confirmed that the error in perception was decreased. Results of both user experiments are publicly-available datasets.

Molecular surfaces provide a suitable way to analyze and to study the evolution and interaction of molecules. The analysis is often concerned with visual identification of binding sites of ligands to a host macromolecule. We present a novel technique that is based on implicit representation, which extracts all potential binding sites and allows an advanced 3D visualization of these sites in the context of the molecule. We utilize implicit function sampling strategy to obtain potential cavity samples and graph algorithms to extract arbitrary cavity components defined by simple graphs. Moreover, we propose how to interactively visualize these graphs in the context of the molecular surface. We also introduce a system of linked views depicting various graph parameters that are used to perform a more elaborative study on created graphs.

Microarray data represents the expression levels of genes for different samples and for different conditions. It has been a central topic in bioinformatics research for a long time already. Researchers try to discover groups of genes that are responsible for specific biological processes. Statistical analysis tools and visualizations have been widely used in the analysis of microarray data. Researchers try to build hypotheses on both the genes and the samples. Therefore, such analyses require the joint exploration of the genes and the samples. However, current methods in interactive visual analysis fail to provide the necessary mechanisms for this joint analysis. In this paper, we propose an interactive visual analysis framework that enables the dual analysis of the samples and the genes through the use of integrated statistical tools. We introduce a set of specialized views and a detailed analysis procedure to describe the utilization of our framework.

In this study, an information theory based framework to automatically construct analytical maps of crowd’s locomotion, called behavior maps, is presented. For these behavior maps, two distinct use cases in crowd simulation domain are introduced; i) automatic camera control ii) behavioral modeling.

Cluster analysis is a popular method for data investigation where data items are structured into groups called clusters. This analysis involves two sequential steps, namely cluster formation and cluster evaluation. In this paper, we propose the tight integration of cluster formation and cluster evaluation in interactive visual analysis in order to overcome the challenges that relate to the black-box nature of clustering algorithms. We present our conceptual framework in the form of an interactive visual environment. In this realization of our framework, we build upon general concepts such as cluster comparison, clustering tendency, cluster stability and cluster coherence. Additionally, we showcase our framework on the cluster analysis of mixed lipid bilayers.

Cluster analysis is a useful method which reveals underlying structures and relations of items after grouping them into clusters. In the case of temporal data, clusters are defined over time intervals where they usually exhibit structural changes. Conventional cluster analysis does not provide sufficient methods to analyze these structural changes, which are, however, crucial in the interpretation and evaluation of temporal clusters. In this paper, we present two novel and interactive visualization techniques that enable users to explore and interpret the structural changes of temporal clusters. We introduce the temporal cluster view, which visualizes the structural quality of a number of temporal clusters, and temporal signatures, which represents the structure of clusters over time. We discuss how these views are utilized to understand the temporal evolution of clusters. We evaluate the proposed techniques in the cluster analysis of mixed lipid bilayers.

Crowds must be simulated believable in terms of their appearance and behavior to improve a virtual environment’s realism. Due to the complex nature of human behavior, realistic behavior of agents in crowd simulations is still a challenging problem. In this paper, we propose a novel behavioral model which builds analytical maps to control agents’ behavior adaptively with agent-crowd interaction formulations. We introduce information theoretical concepts to construct analytical maps automatically. Our model can be integrated into crowd simulators and enhance their behavioral complexity. We made comparative analyses of the presented behavior model with measured crowd data and two agent-based crowd simulators.

Web search query logs contain valuable information which can be utilized for personalization and improvement of search engine performance. The aim in this paper is to cluster users based on their interests, and analyze the temporal dynamics of these clusters. In the proposed approach, we first apply clustering techniques to group similar users with respect to their web searches. Anticipating that the small number of query terms used in search queries would not be sufficient to obtain a proper clustering scheme, we extracted the summary content of the clicked web page from the query log. In this way, we enriched the feature set more efficiently than the content crawling. We also provide preliminary survey results to evaluate clusters. Clusters may change with the user flow from one cluster to the other as time passes. This is due to the fact that users' interests may shift over time. We used statistical methods for the analysis of temporal changes in users' interests. As a case study, we experimented on the query logs of a search engine.

In many application fields, data analysts have to deal with datasets that contain many expressions per item. The effective analysis of such multivariate datasets is dependent on the user's ability to understand both the intrinsic dimensionality of the dataset as well as the distribution of the dependent values with respect to the dimensions. In this paper, we propose a visualization model that enables the joint interactive visual analysis of multivariate datasets with respect to their dimensions as well as with respect to the actual data values. We describe a dual setting of visualization and interaction in items space and in dimensions space. The visualization of items is linked to the visualization of dimensions with brushing and focus+context visualization. With this approach, the user is able to jointly study the structure of the dimensions space as well as the distribution of data items with respect to the dimensions. Even though the proposed visualization model is general, we demonstrate its application in the context of a DNA microarray data analysis.