Research and teaching assistant
Vienna University of Technology
Vienna University of Technology
Vienna University of Technology
Danube University Krems
Ph.D. in Computer Science
Vienna University of Technology
Master in Telecommunication Engineering
Federico II University of Naples, Italy
My main research interest is Visual Analytics (VA), which has beed defined as "the science of analytical reasoning facilitated by interactive visual interfaces" [Thomas and Cook, 2005]. I aim at designing, developing, and validating novel VA techniques to help people analyse complex real-world problems and make sense of big data, by intertwining the immense computing power of machines and the non-replicable perceptual and cognitive capabilities of humans.
In particular, I am interested in VA of large, multivariate, time-oriented, relational data in diverse application domains, such as, for example, social science and medical informatics.
I have been reviewing for several journals (IEEE Transactions on Visualisation and Computer Graphics, Journal of the American Medical Informatics Association, Computer Graphics Forum) and conferences (CHI, EuroVA, EuroVis, InfoVis, IVAPP, PacificVis, SIGGRAPH, SouthCHI, VAST).
MobiGuide (MG) will develop a patient guidance system that integrates hospital and monitoring data into a Personal Health Record (PHR) accessible by patients and care providers and provide personalized secure clinical-guideline-based guidance also outside clinical environments. MG's ubiquity will be achieved by having a Decision Support System (DSS) at the back end, and on the front end by utilizing Body Area Network (BAN) technology and developing a coordinated light-weight DSS that can operate independently. Personalization will be achieved by considering patient preferences and context. Retrospective data analysis will be used to assess compliance and to indicate care pathways shown to be beneficial for certain patient context.
MG will be validated on pre-selected clinical domains with intensive vs. sparse monitoring to demonstrate the generality of the design and assess functionality, feasibility, and impact. MG addresses EU priorities: increasing patient safety, ubiquituous secure access to health care, patient empowerment, developing a common platform for healthcare services, and competitiveness of Europe.
The time is right for MG in view of Europe's vast interest in national PHRs and patient empowerment. MG will leverage this momentum to create a solution that goes beyond local proprietary and stand-alone EMR, DSS, and BAN.
The EXPAND project addresses highly topical challenges of visual analytics methods in the patent data domain. Aiming on a genuinely dynamical framework for visual analysis and knowledge crystallization, radically new concepts and methods have to be developed to unlock the potential of patent network data, which is characterized by its continuous and event based nature, its richness of attributes and its large scale.
While these objectives are ranging high on a scientific and technological research agenda, a user-centred design and evaluation approach will ensure the practical utility and usability of the intended methods and concepts, which will be bundled and evaluated in form of a research prototype.
The advancement and future implementation of the prototype into the business partners IT service portfolio will guarantee the exploitation of results as well as the strengthening of economic advantage in the strategic business intelligence market.
Due to the proliferating capabilities to generate and collect vast amounts of data and information we face the challenge that users and analysts get lost in irrelevant, or otherwise inappropriately processed or presented information. This phenomenon, commonly known as information deluge, overwhelms traditional methods of data analysis such as spreadsheets, ad-hoc queries, or simple visualizations. At the same time, intelligent usage of increasingly available data offers great opportunities to promote technological progress and business success. On this score, Visual Analytics is an emerging research discipline developing methods and technology that make the best possible use of huge information loads in a wide variety of applications. The basic idea is to appropriately combine the strengths of both, computers' and humans' information processing capabilities. To make complex information structures more comprehensible, facilitate new insights, and enable knowledge discovery, methods of visualization, intelligent data analysis, and mining form a symbiosis with the human user via interactive visual interfaces.
The goals of the Centre of Visual Analytics Science and Technology (CVAST) are twofold. The first goal is the integration of the outstanding capabilities of humans in terms of visual information exploration with the enormous processing power of computers to form a powerful knowledge discovery environment. The second goal is to scientifically assess the usability and utility of such discovery environments while bridging the gap between theory and practice for selected application scenarios.
Application scenarios that involve temporal properties are in the focus of our scientific interest. Time - in contrast to other quantitative data dimensions that are usually "flat" - has an inherent structure and distinct characteristics (calendar aspect, natural and social aspects, etc.) which increase its complexity dramatically and demand specialized Visual Analytics methods in order to support proper analysis and visualization. Our expected results will be Innovative, user-oriented, and task-specific Visual Analytics methods and tools with an assessment of their usability and utility. These methods will be used intertwinedly and iteratively to ease the explorative information discovery processes.
Collaboration is the way work gets done in organisations. Therefore networks of different types, functions and compositions have become an inevitable precondition of organisational performance in the modern corporate world. However, these networks are not static, but changing over time. Organisational analysts such as consultants, enterprise analysts, and strategic and human resource managers are seeking for appropriate tools and methods supporting the information discovery processes of these dynamic networks and the underlying data. In particular, interactive visual and analytic methods to explore the dynamic properties of underlying structures are required to provide empirically grounded decision support for organisational analysts.
To meet these demands the proposed project "Visual Enterprise Network Analytics" (VIENA) takes on the method of Dynamic Network Analysis (DNA) whose potential for applied use in enterprises is well known. To unlock this potential also to non-domain experts two associated fields of research are leveraged: namely Visual Analytics and Usability Engineering. A constant focus on usability will assure the development of a user-centred software prototype and an intuitive GUI, while the Visual Analytics components will allow the radically new interactive exploration of dynamic network data sets and the retrieval of relevant information. VIENA facilitates the (semi-)automatic analysis of teams and organisations over different periods of time.
The heterogeneous dynamical data and information sets which get structured, semantically annotated, and visualised by DNA methods get additionally processed using innovative Visual Analytics methods. The aimed integration of selected research prototypes into the industrial partner's business software environment will allow real-user testing and assessments from the application and utility point of view.
In this page I maintain an (almost) updated list of my publications, including downloadable preprints as well as links to definitive published versions on the websites of the respective publishers.
A list including supervised theses / academic works can be automatically generated from the publication database of the Faculty of Informatics of TU Wien here.
Online survey: www.paperviz.org
The increasingly large number of available writings describing technical and scientific progress, calls for advanced analytic tools for their efficient analysis. This is true for many application scenarios in science and industry and for different types of writings, comprising patents and scientific articles. Despite important differences between patents and scientific articles, both have a variety of common characteristics that lead to similar search and analysis tasks. However, the analysis and visualization of these documents is not a trivial task due to the complexity of the documents as well as the large number of possible relations between their multivariate attributes. In this survey, we review interactive analysis and visualization approaches of patents and scientific articles, ranging from exploration tools to sophisticated mining methods. In a bottom-up approach, we categorize them according to two aspects: (a) data type (text, citations, authors, metadata, and combinations thereof), and (b) task (finding and comparing single entities, seeking elementary relations, finding complex patterns, and in particular temporal patterns, and investigating connections between multiple behaviours). Finally, we identify challenges and research directions in this area that ask for future investigations.
Characterizing the problem domain and understanding users' practices and processes are recognized as important steps in order to design and validate visualization, but are often disregarded in practice, also because of their complexity. We introduce the nested work ow model for design and validation of visual analytics, aimed at providing designers with a powerful and expressive modelling tool. This model enables the description of visual analytics processes, at different design levels, in terms of tasks, data, and users, including complex work ow patterns, data and knowledge flows, and collaboration between users. We discuss its application to two visual analytics projects, demonstrating its usefulness for their design and validation.
Several techniques for visualization of dynamic graphs are based on different spatial arrangements of a temporal sequence of node-link diagrams. Many studies in the literature have investigated the importance of maintaining the user's mental map across this temporal sequence, but usually each layout is considered as a static graph drawing and the effect of user interaction is disregarded. We conducted a task-based controlled experiment to assess the effectiveness of two basic interaction techniques: the adjustment of the layout stability and the highlighting of adjacent nodes and edges. We found that generally both interaction techniques increase accuracy, sometimes at the cost of longer completion times, and that the highlighting outclasses the stability adjustment for many tasks except the most complex ones.
A small subset of information visualization and the digital humanities focuses on casual users, that is, users in everyday, non-work contexts. The development of digital cultural heritage collections like Europeana.eu for the general public increases the importance of this user group. This position paper reviews parallel findings from both research fields on casual users, their goals and exploration behavior: Casual users often come without concrete information needs, they rather search for something interesting and engaging. They browse through the information until they find something worth exploring in detail. From these characteristic needs and behavior of casual users we delineate design requirements for public information visualizations in the digital humanities. We discuss how further exchange between these two fields of research can bring forward both fields.
During the last decades, digitization broadened access to cultural heritage collections for public audiences. Large online databases have been prepared for open access with simple search interfaces or visual exploration methods. In this position paper we discuss new challenges arising from these initiatives with regard to casual users. To meet their specific needs, we introduce a novel method for synoptic collection visualization which makes use of parallel space-time cubes to provide multiple spatiotemporal overviews, support free exploration, and to specifically engage casual audiences.
A number of studies have investigated different ways of visualizing uncertainty. However, in the temporal dimension, it is still an open question how to best represent uncertainty, since the special characteristics of time require special visual encodings and may provoke different interpretations. Thus, we have conducted a comprehensive study comparing alternative visual encodings of intervals with uncertain start and end times: gradient plots, violin plots, accumulated probability plots, error bars, centered error bars, and ambiguation. Our results reveal significant differences in error rates and completion time for these different visualization types and different tasks. We recommend using ambiguation - using a lighter color value to represent uncertain regions - or error bars for judging durations and temporal bounds, and gradient plots - using fading color or transparency - for judging probability values.
The advanced visualization of electronic health records (EHRs), supporting a scalable analysis from single patients to cohorts, intertwining patient conditions with executed treatments, and handling the complexity of time-oriented data, is an open challenge of visual analytics for health care. We propose an approach that, according to the knowledge-assisted visualization paradigm, leverages the domain knowledge acquired by clinical experts and formalized into computer-interpretable guidelines, in order to improve the automated analysis, the visualization, and the interactive exploration of EHRs of patient cohorts. In this way, the analyst can get insights about the clinical history of multiple patients and assess the effectiveness of their health care treatments.
Patents, archived as large collections of semi-structured text documents, contain valuable information about historical trends and current states of R&D fields, as well as performances of single inventors and companies. Specific methods are needed to unlock this information and enable its insightful analysis by investors, executives, funding agencies, and policy makers. In this position paper, we propose an approach based on modelling patent repositories as multivariate temporal networks, and examining them by the means of specific visual analytics methods. We illustrate the potential of our approach by discussing two use-cases: the determination of emerging research fields in general and within companies, as well as the identification of inventors characterised by different temporal paths of productivity.
Large multivariate time-oriented networks have been gaining an increasing relevance in different domains. In order to support Visual Analytics processes on this kind of data, appropriate storage and retrieval methods are needed that take into account the scale, dimensionality, and in particular the complex nature of time. We introduce TimeGraph, a data management framework consisting of a data model and two levels of abstraction. TimeGraph captures both the topology of networks and the inherent structure of time into a property graph data structure, and transparently handles them by graph-based operations. TimeGraph aims to be an expressive, easy-to-use and extensible framework, enabling data reduction by selection and aggregation over both the temporal and the topological properties of data, to foster interactive visualization and analysis.
In several application fields, the joint visualization of quantitative data and qualitative abstractions can help analysts make sense of complex time series data by associating precise numeric values with corresponding domain-specific interpretations, such as good, bad, high, low, normal. At the same time, the need to analyse large multivariate time-oriented datasets often calls for keeping visualizations as compact as possible. In this paper, we introduce Qualizon Graphs, a compact visualization that combines quantitative data and qualitative abstractions. It is based on the well known Horizon Graphs, but instead of a predefined number of equally sized bands, it uses as many bands as qualitative categories with corresponding different sizes. In this way, Qualizon Graphs increase the data density of visualized quantitative values and inherently integrate qualitative abstractions. A user study shows that Qualizon Graphs are as fast and accurate as Horizon Graphs for quantitative data, and are an alternative to state-of-the-art visualizations for both quantitative and qualitative data, enabling a trade-off between speed and accuracy.
The analysis of dynamic network data has become an increasingly relevant research issue, showing a high potential for applied use in organizations. To unlock its potential also for the target user group of non-domain experts, we introduce a software prototype, which provides different views on network dynamics, intertwining network analytical measures with options of visual exploration. To demonstrate, how this approach can provide new access to questions of knowledge management and accessibility, a case study of a university department will be discussed. By combining multi-relational data of communication networks with attribute data of individual knowledge domains, we show how essential knowledge and change management issues can be reframed from a social network perspective and further developed towards integrated applications in organizations.
Clinical guidelines provide recommendations in the form of applicable actions in a specific clinical context. Computer Interpretable Guidelines (CIG) aim to achieve guideline integration into clinical practice to increase health care quality. Analyzing the compliance with a CIG can facilitate the implementation and assist in the design of CIGs, but to help medical experts in the detection of patterns in the wealth of the data is a challenging task. We suggest an approach based on visual analytics, intertwining interactive visualization and automated data analysis i.e. analysis of compliance with a CIG. Our solution covers highlighting and abstraction for time-oriented patient parameters, and aggregation of repeatedly missing actions into intervals; in addition valid, invalid, and missing actions are represented visually. Furthermore, we discuss a case study showing how the applied techniques can assist in the detection of interesting patterns.
In recent years, the analysis of dynamic network data has become an increasingly prominent research issue. While several visual analytics techniques with the focus on the examination of temporal evolving networks have been proposed in recent years, their effectiveness and utility for end users need to be further analyzed. When dealing with techniques for dynamic network analysis, which integrate visual, computational, and interactive components, users become easily overwhelmed by the amount of information displayed-even in case of small sized networks. Therefore we evaluated visual analytics techniques for dynamic networks during their development, performing intermediate evaluations by means of mock-up and eye-tracking studies and a final evaluation of the running interactive prototype, tracing three pathways of development in detail: The first one focused on the maintenance of the user s mental map throughout changes of network structure over time, changes caused by user interactions, and changes of analytical perspectives. The second one addresses the avoidance of visual clutter, or at least its moderation. The third pathway of development follows the implications of unexpected user behaviour and multiple problem solving processes. Aside from presenting solutions based on the outcomes of our evaluation, we discuss open and upcoming problems and set out new research questions.
The visual exploration and analysis of time-oriented data in healthcare are important yet challenging tasks. This position paper presents six challenges for Visual Analytics in healthcare: (1) scale and complexity of time-oriented data, (2) intertwining patient condition with treatment processes, (3) scalable analysis from single patients to cohorts, (4) data quality and uncertainty, (5) interaction, user interfaces, and the role of users, and (6) evaluation. Furthermore, it portrays existing and future work by the authors tackling these challenges.
The visualization and analysis of dynamic social networks are challenging problems, demanding the simultaneous consideration of relational and temporal aspects. In order to follow the evolution of a network over time, we need to detect not only which nodes and which links change and when these changes occur, but also the impact they have on their neighbourhood and on the overall relational structure. Aiming to enhance the perception of structural changes at both the micro and the macro level, we introduce the change centrality metric. This novel metric, as well as a set of further metrics we derive from it, enable the pairwise comparison of subsequent states of an evolving network in a discrete-time domain. Demonstrating their exploitation to enrich visualizations, we show how these change metrics support the visual analysis of network dynamics.
A well-designed visualization of dynamic networks has to support the analysis of both temporal and relational fea- tures at once. In particular to solve complex synoptic tasks, the users need to understand the topological structure of the network, its evolution over time, and possible interde- pendencies. In this paper, we introduce the application of the vertigo zoom interaction technique, derived from film- making, to information visualizations. When applied to a two-and-a-half-dimensional view, this interaction technique enables smooth transitions between the relational perspec- tive (node-link diagrams and scatter plots) and the time per- spective (trajectories and line charts), supporting a seamless visual analysis and preserving the user´s mental map.
Despite its well-known potential for applied use in organizations, social network analysis seems to fail relevant business analytical requirements in the areas of organizational change and software usability for non-expert users like managers and consultants. This position paper takes on this challenge by outlining a strategy of user-driven software development which aims to shift analytical procedures from the numerical to the visual realm. As network dynamics can be visualized using various methods, a comparative analysis of their respective strengths and weaknesses lays the basis for the suggested integration of additional visual methods into network exploration and interpretation procedures.
The visualization and analysis of dynamic networks have become increasingly important in several fields, for instance sociology or economics. The dynamic and multi-relational nature of this data poses the challenge of understanding both its topological structure and how it changes over time. In this paper we propose a visual analytics approach for analyzing dynamic networks that integrates: a dynamic layout with user-controlled trade-off between stability and consistency; three temporal views based on different combinations of node-link diagrams (layer superimposition, layer juxtaposition, and two-and-a-half dimensional view); the visualization of social network analysis metrics; and specific interaction techniques for tracking node trajectories and node connectivity over time. This integration of visual, interactive, and automatic methods supports the multifaceted analysis of dynamically changing networks.
Semester hours: 2.0; Credits: 3.0; Type: VO
Semester hours: 1.0; Credits: 1.5; Type: UE
Semester hours: 6.0; Credits: 12.0; Type: PR
Semester hours: 6.0; Credits: 9.0; Type: PR
Semester hours: 4.0; Credits: 2.0; Type: PR
Semester hours: 4.0; Credits: 6.0; Type: PR
Semester hours: 4.0; Credits: 2.0; Type: PR
Semester hours: 2.0; Credits: 3.0; Type: SE
Semester hours: 2.0; Credits: 3.0; Type: SE