• Publikationen
Show publication details

Ceneda, Davide; Gschwandtner, Theresia; May, Thorsten; Miksch, Silvia; Schulz, Hans-Jörg; Streit, Marc; Tominski, Christian

Characterizing Guidance in Visual Analytics

2017

IEEE Transactions on Visualization and Computer Graphics, Vol.23 (2017), 1, pp. 111-120

IEEE Conference on Visual Analytics Science and Technology (VAST) <11, 2016, Baltimore, USA>

Visual analytics (VA) is typically applied in scenarios where complex data has to be analyzed. Unfortunately, there is a natural correlation between the complexity of the data and the complexity of the tools to study them. An adverse effect of complicated tools is that analytical goals are more difficult to reach. Therefore, it makes sense to consider methods that guide or assist users in the visual analysis process. Several such methods already exist in the literature, yet we are lacking a general model that facilitates in-depth reasoning about guidance. We establish such a model by extending van Wijk's model of visualization with the fundamental components of guidance. Guidance is defined as a process that gradually narrows the gap that hinders effective continuation of the data analysis. We describe diverse inputs based on which guidance can be generated and discuss different degrees of guidance and means to incorporate guidance into VA tools. We use existing guidance approaches from the literature to illustrate the various aspects of our model. As a conclusion, we identify research challenges and suggest directions for future studies. With our work we take a necessary step to pave the way to a systematic development of guidance techniques that effectively support users in the context of VA.

Show publication details

Fellner, Dieter W.; Baier, Konrad; Ackeren, Janine van; Alexandrin, Max; Barth, Anna; Bockholt, Ulrich; Kopold, Franziska; Löwer, Chris; May, Thorsten; Peters, Wiebke; Wehner, Detlef; Gollnast, Anja; Bumke, Carina

Jahresbericht 2016: Fraunhofer-Institut für Graphische Datenverarbeitung IGD

2017

Darmstadt, 2017

Das Fraunhofer IGD hat seine Forschungsaktivitäten vor Kurzem in vier Leitthemen gebündelt, welche die Basis seiner Arbeit bilden und verschiedene Themen abteilungsübergreifend miteinander verknüpfen. Eines dieser Leitthemen ist "Visual Computing as a Service - Die Plattform für angewandtes Visual Computing". Die Basis dieser universellen Plattform für Visual-Computing Lösungen ist gelegt und wird kontinuierlich erweitert. Dieser technologische Ansatz bildet die Grundlage für die weiteren Leitthemen. In der "Individuellen Gesundheit - Digitale Lösungen für das Gesundheitswesen" werden die Daten betrachtet, die in der personalisierten Medizin anfallen - mithilfe der Visual-Computing-Technologien des Instituts. Im Leitthema "Intelligente Stadt - Innovativ, digital und nachhaltig" ist die Fragestellung, wie man den Lebenszyklus urbaner Prozesse unterstützen kann. Und im Leitthema "Digitalisierte Arbeit - Der Mensch in der Industrie 4.0" geht es erster Linie um die Unterstützung des Menschen in der durch die Digitalisierung veränderten Produktion.

Show publication details

Reynolds, Steven Lamarr; Kuijper, Arjan (Betreuer); May, Thorsten (Betreuer)

Visualisierung von zeitlichen, thematischen Änderungen in einem Dokumentenkorpus

2017

Darmstadt, TU, Bachelor Thesis, 2017

Mehrere Dokumente in einem Korpus können mit Themen beschrieben werden. Diese Themen bestehen aus Termen die sich aus den Wörtern der Dokumente bilden. Das Thema der vorliegenden Bachelorarbeit beschäftigt sich mit der Visualisierung der Entwicklung dieser Themen über einen Zeitstrahl. Über mehrere Zeitabschnitte hinweg werden Themen aus Dokumenten generiert. In einer Visualisierung wird die Entwicklung der Themen über diese Zeitabschnitte dargestellt. Dazu wird die Ähnlichkeit zwischen Themen berrechnet und angezeigt. Die Themen werden sortiert um die Visualisierung möglichst visuell und ansprechend darzustellen. Damit können Entwicklungen der Themen betrachtet werden. Das System erlaubt ein späteres austauschen des verwendeten Topic Modeling Verfahrens und ein schnellen Import von Dokumentenkorpera. Die Visualisierung ist unabhängig von dem Topic Modeling Verfahren. Zusätzlich wird mittels Interaktionstechniken eine Visual Analytics Anwendung umgesetzt.

Show publication details

Ruddle, Roy A.; Bernard, Jürgen; May, Thorsten; Lücke-Tieke, Hendrik; Kohlhammer, Jörn

Methods and a Research Agenda for the Evaluation of Event Sequence Visualization Techniques

2016

IEEE Computer Society Visualization and Graphics Technical Committee (VGTC): IEEE VIS 2016 Workshop on Temporal and Sequential Event Analysis : The Event Event: Temporal & Sequential Event Analysis [online]. [cited 01 February 2017] Available from: [http://eventevent.github.io/], 2016, 4 p.

IEEE VIS 2016 Workshop on Temporal & Sequential Event Analysis <2016, Baltimore, USA>

The present paper asks how can visualization help data scientists make sense of event sequences, and makes three main contributions. The first is a research agenda, which we divide into methods for presentation, interaction & computation, and scale-up. Second, we introduce the concept of Event Maps to help with scale-up, and illustrate coarse-, medium- and fine-grained Event Maps with electronic health record (EHR) data for prostate cancer. Third, in an experiment we investigated participants' ability to judge the similarity of event sequences. Contrary to previous research into categorical data, color and shape were better than position for encoding event type. However, even with simple sequences (5 events of 3 types in the target sequence), participants only got 88% correct despite averaging 7.4 seconds to respond. This indicates that simple visualization techniques are not effective.

Show publication details

Beuth, Marcel; Kuijper, Arjan (Betreuer); May, Thorsten (Betreuer)

Visualization of General Graphs with Metro Maps

2016

Darmstadt, TU, Bachelor Thesis, 2016

In this work an algorithm is developed that generates a Metro Map layout for general graphs that is not based on a geographical background. Metro Maps have the great advantage of good and intuitive readability. The developed algorithm is based on the work of Stott et al. The special feature in this work is that the visualized graph is a partial graph of a more complex graph. This embedding has influence to the visualization of the Metro Map.

Show publication details

Bernard, Jürgen; Sessler, David; Bannach, Andreas; May, Thorsten; Kohlhammer, Jörn

A Visual Active Learning System for the Assessment of Patient Well-Being in Prostate Cancer Research

2015

Gschwandtner, Theresia (Conference Chair) et al.: Proceedings of the 2015 Workshop on Visual Analytics in Healthcare. New York: ACM, 2015, Art. 1, 8 p.

Workshop in Visual Analytics in Healthcare (VAHC) <2015, Chicago, IL, USA>

The assessment of patient well-being is highly relevant for the early detection of diseases, for assessing the risks of therapies, or for evaluating therapy outcomes. The knowledge to assess a patient's well-being is actually tacit knowledge and thus, can only be used by the physicians themselves. The rationale of this research approach is to use visual interfaces to capture the mental models of experts and make them available more explicitly. We present a visual active learning system that enables physicians to label the well-being state of patient histories su ering prostate cancer. The labeled instances are iteratively learned in an active learning approach. In addition, the system provides models and visual interfaces for a) estimating the number of patients needed for learning, b) suggesting meaningful learning candidates and c) visual feedback on test candidates. We present the results of two evaluation strategies that prove the validity of the applied model. In a representative real-world use case, we learned the feedback of physicians on a data collection of more than 16.000 prostate cancer histories.

Show publication details

Bernard, Jürgen; Sessler, David; May, Thorsten; Schlomm, Thorsten; Pehrke, Dirk; Kohlhammer, Jörn

A Visual-Interactive System for Prostate Cancer Cohort Analysis

2015

IEEE Computer Graphics and Applications, Vol.35 (2015), 3, pp. 44-55

Data-centered research is becoming increasingly important in prostate cancer research, where a long-term goal is a sound prognosis prior to surgery. The proposed visual-interactive system, developed in close collaboration with medical researchers, helps physicians efficiently and effectively visualize single and multiple patient histories at a glance.

Show publication details

Gschwandtner, Theresia; Schumann, Heidrun; Bernard, Jürgen; May, Thorsten; Bögl, Markus; Miksch, Silvia; Kohlhammer, Jörn; Röhlig, Martin; Alsallakh, Bilal

Enhancing Time Series Segmentation and Labeling Through the Knowledge Generation Model: Poster presented at the Eurographics Conference on Visualization (EuroVis)

2015

2015

Eurographics Conference on Visualization (EuroVis) <17, 2015, Cagliari, Sardinia, Italy>

Segmentation and labeling of different activities in multivariate time series data is an important task in many domains. There is a multitude of automatic segmentation and labeling methods available, which are designed to handle different situations. These methods can be used with multiple parametrizations, which leads to an overwhelming amount of options to choose from. To this end, we present a conceptual design of a Visual Analytics framework (1) to select appropriate segmentation and labeling methods with appropriate parametrizations, (2) to analyze the (multiple) results, (3) to understand different kinds and origins of uncertainties in these results, and (4) to reason which methods and which parametrizations yield stable results and fine-tune these configurations if necessary.

Show publication details

Maier, Sebastian; Kühnel, Hannes; May, Thorsten; Kuijper, Arjan

Visual Interactive Process Monitoring

2015

Yamamoto, Sakae (Ed.): Human Interface and the Management of Information. Proceedings Part I : Information and Knowledge Design. Springer International Publishing, 2015. (Lecture Notes in Computer Science (LNCS) 9172), pp. 74-85

International Conference on Human Interface and the Management of Information (HIMI) <2015, Los Angeles, CA, USA>

Sensor data has been coined the oil of the 21st century. We present a technique for the visual analysis of multivariate sensor event log data. This technique tackles two challenges: Firstly, in a complex process the relation of causes and effects is often masked by indirections. Secondly, the metrics to measure success might be different from the measures that identify causes. Thus, our approach does not require that all sensor data is equal. Our techniques combines automated and interactive grouping to identify candidate sets sharing properties relevant for cause and effect analysis. Interactive visual probes offer immediate information on the statistical relevance of an identified connection.

Show publication details

Ruppert, Tobias; Bernard, Jürgen; Lücke-Tieke, Hendrik; May, Thorsten; Kohlhammer, Jörn

Visual-Interactive Text Analysis to Support Political Decision Making - From Sentiments to Arguments to Policies

2015

Bertini, Enrico (Ed.) et al.: EuroVA 2015 : EuroVis Workshop on Visual Analytics. Goslar: Eurographics Association, 2015, pp. 37-41

International EuroVis Workshop on Visual Analytics (EuroVA) <6, 2015, Cagliari, Sardinia, Italy>

Political decision making involves the evaluation of alternative solutions (so called policy models) to a given societal problem and the selection of the most promising one. Large amounts of textual information to be considered in decision making processes can be found on the web. This includes general information about policy models, individual arguments in favor or against these policies, and public opinions. Monitoring large text collections and extracting the relevant information is time consuming. In this approach we present a visual analytics system that supports users in assessing the results of automatic text analysis methods. The methods extract text segments from large document collections and associate them with predefined policy domains, policy models, and policy arguments. Moreover, sentiment analysis is applied on the text segments. Visualization techniques provide non-IT experts an intuitive access to the results. With the system, users can monitor public debates on the web. In addition, we analyze concepts that enable the user to give visual-interactive feedback on the text analysis results. This direct user feedback can help to improve the accuracy of individual text analysis modules and the credibility of the overall text analysis process. The system was tested with real users from the political decision making domain.

Show publication details

Steiger, Martin; Bernard, Jürgen; May, Thorsten; Kohlhammer, Jörn

A Survey of Direction-Preserving Layout Strategies

2014

Gutierrez, Diego (Ed.): Proceedings of the 30th Spring Conference on Computer Graphics : SCCG 2014. New York: ACM, 2014, pp. 21-28

Spring Conference on Computer Graphics (SCCG) <30, 2014, Smolenice, Slovakia>

In this paper we analyze different layout algorithms that preserve relative directions in geo-referenced networks. This is an important criterion for many sensor networks such as the electric grid and other supply networks, because it enables the user to match the geographic setting with the drawing on the screen. Even today, the layouts of these networks are often created manually. This is due to the requirement that these layouts must respect geographic references but should still be easy to read and understand. The range of available automatic algorithms spans from general graph layouts over schematic maps to semi-realistic drawings. At first sight, schematics seem to be a promising compromise between geographic correctness and readability. The former property exploits the mental map of the user while the latter makes it easier for the user to learn about the network structure. We investigate different algorithms for such maps together with different visualization techniques. In particular, the group of octi-linear layouts is prominent in handcrafted subway maps. These algorithms have been used extensively to generate drawings for subway maps. Also known as Metro Map layouts, only horizontal, vertical and diagonal directions are allowed. This increases flexibility and makes the resulting layout look similar to the well-known subway maps of large cities. The key difference to general graph layout algorithms is that geographic relations are respected in terms of relative directions. However, it is not clear, whether this metaphor can be transferred from metro maps to other domains. We discuss applicability of these different approaches for geo-based networks in general with the electric grid as a use-case scenario

Show publication details

Ruppert, Tobias; Bernard, Jürgen; May, Thorsten; Kohlhammer, Jörn

Combining Computational Models and Interactive Visualization to Support Rational Decision Making

2014

Bebis, George (Ed.) et al.: Advances in Visual Computing. 10th International Symposium, ISVC 2014 : Proceedings, Part I. Berlin, Heidelberg, New York: Springer, 2014. (Lecture Notes in Computer Science (LNCS) 8887), pp. 345-356

International Symposium on Visual Computing (ISVC) <10, 2014, Las Vegas, NV, USA>

Decision making is a complex process consisting of several consecutive steps. Before converting a decision into effective action the problem to be tackled needs to be analyzed, alternative solutions need to be developed, and the best solution needs to be picked. In many cases computational models support decision makers in this process. Therefore, providing an intuitive access to these model-driven techniques is crucial. In this approach, we introduce a decision support system that provides visual-interactive access to three computational models - a simulation model, an optimization model, and an opinion mining model - covering different aspects of decision making. For each model our decision support system realizes the visual access to the model, an in-depth analysis of the generated solutions, and the comparison of alternative solutions. Finally, we evaluate the usefulness and the usability of our system in a use case in the field of public policy making.

Show publication details

Steiger, Martin; Lücke-Tieke, Hendrik; May, Thorsten; Kuijper, Arjan; Kohlhammer, Jörn

Deterministic Local Layouts through High-Dimensional Layout Stitching

2014

Kurosu, Masaaki (Ed.): Human-Computer Interaction: Part 1 : Theories, Methods, and Tools. HCI International 2014. Berlin, Heidelberg, New York: Springer, 2014. (Lecture Notes in Computer Science (LNCS) 8510), pp. 643-651

International Conference on Human-Computer Interaction (HCII) <16, 2014, Heraklion, Crete, Greece>

In this paper we present a layout technique for dynamic views of large static graphs. It aims to minimize changes between two consecutive frames and most importantly, it is deterministic. First, a set of small layout patches is pre-computed. Then, depending on the users view focus, a subset of these patches is selected and connected to generate the final layout. In contrast to the state-of-the-art approach that operates in the 2D screen space only, we perform this process in high-dimensional space before projecting the results into the 2D plane. This gives additional degrees of freedom and consequently a smoother transition process between two consecutive frames. Whenever the user visits an area of the graph for a second time, the layout will still look the same. This enables the user to recognize areas that have already been explored and thus preserve the mental map.

Show publication details

May, Thorsten; Nazemi, Kawa; Kohlhammer, Jörn

From Raw Data to Rich Visualization: Combining Visual Search with Data Analysis

2014

Wahlster, Wolfgang (Ed.) et al.: Towards the Internet of Services: The THESEUS Research Program. Berlin, Heidelberg, New York: Springer, 2014. (Cognitive Technologies), pp. 203-209

Visual analytics is an interdisciplinary field of research at the boundary between data mining, statistics and visualization. Patterns and relations in the data complement a semantic representation of knowledge on a lower level of abstraction. One important goal of visual analytics is to find relations hidden in vast amounts of data, which can be turned into useful knowledge. Analysis needs to be "visual", because human's visual cognitive abilities are important for the identification and refinement of the analytical process. Further the results of the analysis have to be presented in a way to match the user's perspective on the proposed task. However, typical users are not experts in statistics or data mining. The challenge of visual analytics is to keep domain experts in charge of the analytical process while reducing the workload due to the complexity of the techniques. While search and analysis usually arc mentioned in different contexts, they are highly interdependent processes. In fact, every exploratory analysis is a search for new knowledge. In turn, this knowledge can be used to refine future searches by introducing new concepts or relations to draw from. This article will show how automated and visual methods can be combined to connect knowledge artifacts on multiple levels of abstraction.

Show publication details

Steiger, Martin; May, Thorsten; Kohlhammer, Jörn

Stable Incremental Layouts for Dynamic Graph Visualizations

2014

IADIS International Journal on Computer Science and Information Systems, Vol.8 (2014), 2, pp. 12-26

In this paper we present a set of extension techniques to stabilize interactive dynamic graph layout algorithms. It works with different existing Focus & Context methods. We first deal with the initial placement of newly inserted nodes to mitigate acting forces in the layout algorithm. Then, their influence on the existing layout is gradually increased to create a smooth transition between the old and the new layout. To complement this approach we use a look-ahead strategy that integrates additional nodes in the layout to stabilize the layout even more. Our approach is validated using both quantitative and qualitative measures.

Show publication details

Steiger, Martin; Bernard, Jürgen; Mittelstädt, Sebastian; Lücke-Tieke, Hendrik; Keim, Daniel A.; May, Thorsten; Kohlhammer, Jörn

Visual Analysis of Time-Series Similarities for Anomaly Detection in Sensor Networks

2014

Computer Graphics Forum, Vol.33 (2014), 3, pp. 401-410

Eurographics Conference on Visualization (EuroVis) <16, 2014, Swansea, Wales, UK>

We present a system to analyze time-series data in sensor networks. Our approach supports exploratory tasks for the comparison of univariate, geo-referenced sensor data, in particular for anomaly detection. We split the recordings into fixed-length patterns and show them in order to compare them over time and space using two linked views. Apart from geo-based comparison across sensors we also support different temporal patterns to discover seasonal effects, anomalies and periodicities. The methods we use are best practices in the information visualization domain. They cover the daily, the weekly and seasonal and patterns of the data. Daily patterns can be analyzed in a clustering-based view, weekly patterns in a calendar-based view and seasonal patterns in a projection-based view. The connectivity of the sensors can be analyzed through a dedicated topological network view. We assist the domain expert with interaction techniques to make the results understandable. As a result, the user can identify and analyze erroneous and suspicious measurements in the network. A case study with a domain expert verified the usefulness of our approach.

Show publication details

Kühnel, Hannes; Kuijper, Arjan (Betreuer); Maier, Sebastian (Betreuer); May, Thorsten (Betreuer)

Visual Interactive Process Monitoring

2014

Darmstadt, TU, Bachelor Thesis, 2014

In this work, a visual analytics tool is developed which enables automatic analysis of processes, which are defined by the user, of monitoring systems. This tool provides reports of the performance and sequence of process steps and allows a more detailed analysis of user interactions. Based on technologies of visual analytics and information visualization, data sets can be analyzed and presented visually for deeper analysis. This includes a directed node-link diagram, that represents the process model and diagrams for metadata such as the performance of each step. The objective is to be able to handle even large amounts of data and to compare a state of the process with the desired process state. The incoming data are event data consisting of logs from events which represent process steps used in a process containing a timestamp and other information like e.g. money or duration. The problem which has to be solved is to handle the information extracted from event data and to find complex, direct and indirect relations between process steps by giving an overview using a node-link diagram with weighted edges to represent the transitions from one process step to another and providing interaction techniques. Additional the information has to be analyzed according to chosen indicators (e.g. duration or money) and to compare different types of process runs with each other, e.g. separating complete and incomplete runs, to possibly find these relations which can be indicators for an incomplete run. As a result the tool gives the ability to choose different runs to compare them and simultaneously compares the chosen indicators. Furthermore, the tool was evaluated to find out, if it is possible to solve this problems by using this tool.

Show publication details

Bernard, Jürgen; Steiger, Martin; Widmer, Sven; Lücke-Tieke, Hendrik; May, Thorsten; Kohlhammer, Jörn

Visual-interactive Exploration of Interesting Multivariate Relations in Mixed Research Data Sets

2014

Computer Graphics Forum, Vol.33 (2014), 3, pp. 291-300

Eurographics Conference on Visualization (EuroVis) <16, 2014, Swansea, Wales, UK>

The analysis of research data plays a key role in data-driven areas of science. Varieties of mixed research data sets exist and scientists aim to derive or validate hypotheses to find undiscovered knowledge. Many analysis techniques identify relations of an entire dataset only. This may level the characteristic behavior of different subgroups in the data. Like automatic subspace clustering, we aim at identifying interesting subgroups and attribute sets. We present a visual-interactive system that supports scientists to explore interesting relations between aggregated bins of multivariate attributes in mixed data sets. The abstraction of data to bins enables the application of statistical dependency tests as the measure of interestingness. An overview matrix view shows all attributes, ranked with respect to the interestingness of bins. Complementary, a node-link view reveals multivariate bin relations by positioning dependent bins close to each other. The system supports information drill-down based on both expert knowledge and algorithmic support. Finally, visual-interactive subset clustering assigns multivariate bin relations to groups. A list-based cluster result representation enables the scientist to communicate multivariate findings at a glance. We demonstrate the applicability of the system with two case studies from the earth observation domain and the prostate cancer research domain. In both cases, the system enabled us to identify the most interesting multivariate bin relations, to validate already published results, and, moreover, to discover unexpected relations.

Show publication details

Bernard, Jürgen; Wilhelm, Nils; Krüger, Björn; May, Thorsten; Schreck, Tobias; Kohlhammer, Jörn

MotionExplorer: Exploratory Search in Human Motion Capture Data Based on Hierarchical Aggregation

2013

IEEE Transactions on Visualization and Computer Graphics, Vol.19 (2013), 12, pp. 2257-2266

IEEE Symposium on Visual Analytics Science and Technology (VAST) <8, 2013, Atlanta, GA, USA>

We present MotionExplorer, an exploratory search and analysis system for sequences of human motion in large motion capture data collections. This special type of multivariate time series data is relevant in many research fields including medicine, sports and animation. Key tasks in working with motion data include analysis of motion states and transitions, and synthesis of motion vectors by interpolation and combination. In the practice of research and application of human motion data, challenges exist in providing visual summaries and drill-down functionality for handling large motion data collections. We find that this domain can benefit from appropriate visual retrieval and analysis support to handle these tasks in presence of large motion data. To address this need, we developed MotionExplorer together with domain experts as an exploratory search system based on interactive aggregation and visualization of motion states as a basis for data navigation, exploration, and search. Based on an overview-first type visualization, users are able to search for interesting sub-sequences of motion based on a query-by-example metaphor, and explore search results by details on demand. We developed MotionExplorer in close collaboration with the targeted users who are researchers working on human motion synthesis and analysis, including a summative field study. Additionally, we conducted a laboratory design study to substantially improve MotionExplorer towards an intuitive, usable and robust design. MotionExplorer enables the search in human motion capture data with only a few mouse clicks. The researchers unanimously confirm that the system can efficiently support their work.

Show publication details

Steiger, Martin; May, Thorsten; Davey, James; Kohlhammer, Jörn

Smart Grid Monitoring through Visual Analysis

2013

IEEE Power & Energy Society (PES): 2013 4th IEEE PES Innovative Smart Grid Technologies Europe : IEEE PES ISGT Europe 2013. Piscataway: IEEE, 2013, 5 p.

IEEE/PES Innovative Smart Grid Technologies Europe (ISGT Europe) <4, 2013, Copenhagen, Denmark>

In this paper we present a visualization system for the real-time monitoring of Smart Grids. In particular, it supports the control room operators of electric grids with large amounts of distributed power generation. As measurements are continuously read from connected Smart Meters, an expert system performs a classification of these events to relieve the operator from the manual inspection of irrelevant and trivial information items. A coordinated, multiple view environment displays the electric grid model which is annotated with status indicators. This gives the operator both a reasonable overview on the entire system, but also enables him to inspect specific parts in detail. Other sources of information such as ICT coverage and weather conditions are included to provide additional context. The visualization system thus provides monitoring and control support for operators to keep the grid in a stable condition.

Show publication details

Steiger, Martin; May, Thorsten; Kohlhammer, Jörn

Stable Incremental Layouts for Dynamic Graph Visualizations

2013

Xiao, Yingcai (Ed.): Proceedings of the IADIS International Conference Computer Graphics, Visualization, Computer Vision and Image Processing : CGVCVIP 2013. IADIS Press, 2013, pp. 91-98

IADIS International Conference Computer Graphics, Visualization, Computer Vision and Image Processing (CGVCVIP) <2013, Prague, Czech Republic>

In this paper we present a set of extension techniques to stabilize interactive dynamic graph layout algorithms. It works with different existing Focus & Context methods. We first deal with the initial placement of newly inserted nodes to mitigate acting forces in the layout algorithm. Then, their influence on the existing layout is gradually increased to create a smooth transition between the old and the new layout. To complement this approach we use a look-ahead strategy that integrates additional nodes in the layout to stabilize the layout even more.

Show publication details

Steiger, Martin; Lücke-Tieke, Hendrik; May, Thorsten; Kuijper, Arjan; Kohlhammer, Jörn

Using Layout Stitching to Create Deterministic Local Graph Layouts

2013

Oliveira, Manuel M. (Ed.) et al.: WSCG 2013. Full Papers. Proceedings : The 21st International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision. Plzen: University of West Bohemia, 2013, pp. 1-9

International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision (WSCG) <21, 2013, Plzen, Czech Republic>

Dynamic graph layouts are often used to position nodes in local views of large graphs. These layouts can be optimized to minimize changes when navigating to other parts of the graph. Dynamic graph layout techniques do not, however, guarantee that a local layout is recognizable when the user visits the same area twice. In this paper we present a method to create stable and deterministic layouts of dynamic views of large graphs. It is based on a well-known panorama-stitching algorithm from the image processing domain. Given a set of overlapping photographs it creates a larger panorama that combines the original images. In analogy to that our algorithm stitches pre-computed layouts of subgraphs to form a larger, single layout. This deterministic approach makes structures and node locations persistent which creates identical visual representations of the graph. This enables the user to recognize previously encountered parts and to decide whether a certain part of a dataset has already been explored before or not.

Show publication details

Steiger, Martin; May, Thorsten; Davey, James; Kohlhammer, Jörn

Visual Analysis of Expert Systems for Smart Grid Monitoring

2013

Pohl, Margit (Ed.) et al.: EuroVA 2013 : International EuroVis Workshop on Visual Analytics. Goslar: Eurographics Association, 2013, pp. 43-47

International EuroVis Workshop on Visual Analytics (EuroVA) <4, 2013, Leipzig, Germany>

This application paper introduces the combination of an expert system with a visualization system, specifically designed for SmartGrid control rooms. It supports operators in the efficient monitoring of electric grids with a focus on distributed, small-scale power generation. A rule-based expert system filters the stream of a large amount of incoming events, searching for potential problems. A multiple, coordinated view environment provides the required situational awareness for the user and presents the analysis results in a details-on-demand manner. Being a critical infrastructure, the electric grid is highly sensitive and any modification must be well justified. Our visualization system therefore also provides insight into the expert system enabling the user to validate and verify the expert system's analysis process. This provides the required decision support to assist the operator in keeping the grid in stable operating condition.

Show publication details

Davey, James; Hutter, Marco; May, Thorsten; Kohlhammer, Jörn

Combining Linked Matrix Visualizations with Multi-Dimensional Clustering to Detect Cybercrime: Poster presented at the International Workshop on Visual Analytics (EuroVA)

2012

Darmstadt, 2012

International Workshop on Visual Analytics (EuroVA) <3, 2012, Vienna, Austria>

Our goal is to find a useful multi-dimensional clustering (MDC) of a data set from the network security domain. Algorithms from the Multi-Criteria Decision Analysis (MCDA) domain are used for the calculation of fused similarity matrices, in which the information from several dimensions are combined. However, the results of MCDA algorithms are often difficult to interpret. We present a linked, interactive, matrix-based visualization which simplifies the comparison of clusters and increases the understanding of MCDA results.

Show publication details

May, Thorsten; Fellner, Dieter W. (Betreuer); Hauser, Helwig (Betreuer)

Modelle und Methoden für die Kopplung automatischer und visuell-interaktiver Verfahren für die Datenanalyse

2012

Darmstadt, TU, Diss., 2011

In dieser Arbeit werden neue Kopplungsvarianten von Visualisierungstechniken und Data- Mining-Verfahren für die Datenanalyse vorgestellt. Die betrachteten Teilschritte der Analyse sind hierbei die Suche von Mustern, deren Transformation in formale Modelle und die Überprüfung dieser Modelle. Die Suche nach Mustern und deren Modellierung ist in diesen Technologien bisher nicht getrennt. In diesem Konzept werden die Aufgaben neu aufgeteilt: Die Suche und Identifizierung von Mustern übernimmt der Mensch mit Hilfe von interaktiven Visualisierungstechniken. Die Transformation der gefundenen Muster wird durch Data-Mining Verfahren übernommen. Diese Aufteilung nutzt die spezifischen Stärken von Mensch und Maschine besser. Der Mensch trägt mit seinem flexiblen und robusten Wahrnehmungssystem zu dieser Aufgabe bei; er wird gleichzeitig von der kognitiv anspruchsvolleren Aufgabe der Transformation der Muster entlastet. Die Maschine kann auch insbesondere komplexe Muster leichter in formale Modelle umformen. Wenn der Mensch die Interpretation ohne Hilfsmittel oder Übung durchführen muss, können selten mehr als zwei bis dreidimensionale Muster interpretiert werden. In der Arbeit wird gezeigt, dass die Muster zehn- und mehrdimensionaler Zusammenhänge nicht nur dargestellt und wahrgenommen, sondern auch genutzt werden können. Durch die Aufteilung kann auch jene Schwäche automatischer Verfahren kompensiert werden, dass sie nur solche Muster finden, für deren Suche sie entwickelt wurden. Bei der Konstruktion der Modelle eliminiert die interaktive Vorgabe Mehrdeutigkeiten in den Mustern. Zur Überprüfung des konstruierten Modells wird daraus wieder ein Muster generiert, das mit den Mustern der Originaldaten visuell verglichen wird. Dies liefert qualitative Information über die Art des Modellierungsfehlers und die prinzipielle Eignung des automatischen Verfahrens, die quantitative Gütemaße allein nicht bereitstellen können. Zusammengenommen beschreibt das Konzept zwei Verbindungen zwischen visuell-interaktiven Techniken und automatischen Techniken in entgegengesetzte, sich ergänzende Richtungen. Mustererkennung, Interaktion und automatische Modellierung beschreiben den Weg von Muster zum Modell. Simulation, Feedback und visueller Abgleich beschreiben den Weg vom Modell zurück zum Muster. Das Modell der Daten wird in einem zyklischen, iterativen Prozess konstruiert, stetig überprüft und verfeinert. Um diese beiden Varianten der Kopplung von anderen, existierenden Verfahren abgrenzen zu können, wurde das allgemeine Modell des Visual-Analytics-Prozess verfeinert. Unabhängig von spezifischen Techniken wurden acht verschiedene Kopplungsvarianten identifiziert. Dabei werden sowohl bekannte Ansätze in die Systematik eingeordnet, als auch solche, die bisher nicht oder kaum in der Literatur umgesetzt werden.

Show publication details

Lücke-Tieke, Hendrik; May, Thorsten (Betreuer); Kuijper, Arjan (Betreuer)

Stabiles Graph-Layout Basierend auf Lokalen Layouts

2012

Darmstadt, TU, Bachelor Thesis, 2012

Das Ziel einer Datenanalyse ist die Gewinnung von Wissen. Neben der vollautomatischen Analyse ist die interaktive Analyse dieser Daten ein vielversprechender Ansatz, denn sie ermöglicht es, die Vorteile des Menschen gegenüber rein maschinengestützter Verarbeitung zu nutzen. Vorteile wie robuste Wahrnehmung, visuelle Intelligenz und die Fähigkeit zur Entscheidung auf Basis ungenauer Informationen können genutzt werden, um den Wissensgewinn zu beschleunigen. Dies ist allerdings nur nutzbar, wenn umgekehrt der Analyst bei der Entwicklung und Evaluation von Hypothesen unterstützt wird. Eine Form der Unterstützung bietet die grafische Visualisierung von Netzwerken als Diagramme. Diese sogenannten Node-Link-Diagramme stellen Zusammenhänge zwischen Datenpunkten (Nodes) durch Linien (Links) dar, die diese miteinander verbinden. Die Effektivität einer visuellen Analyse mit Hilfe von Node-Link-Diagrammen wird 2 Faktoren stark beeinflusst: die Nachvollziehbarkeit von Zusammenhängen und die Übersichtlichkeit der Visualisierung. Dabei kann die Übersichtlichkeit durch die Menge der dargestellten Daten gesteuert werden. Die Nachvollziehbarkeit von Zusammenhängen hängt hingegen von der Positionierung der Datenpunkte ab. Viele Visualisierungsverfahren für Node-Link-Diagramme betrachten den gesamten Datensatz und erzeugen daraus eine Gesamtdarstellung, ein sogenanntes globales Layout. Dieses globale Layout erleichtert dem Betrachter die Orientierung, denn jeder Datenpunkt befindet sich immer an der gleichen Position. Ein Analyst kann sich so auf den Lageplan in seinem Kopf verlassen, die Visualisierung ist in diesem Sinne stabil. Dies ist allerdings ein Kompromiss auf Kosten der Genauigkeit lokaler Zusammenhänge. Doch genau diese Zusammenhänge können relevant sein und bedürfen daher einer gesonderten Betrachtung ohne Abhängigkeiten oder Einflüsse des restlichen Graphen. Ein solcher, unabhängig berechneter Ausschnitt wird als lokales Layout bezeichnet. Mit Unterstützung von interaktiven Explorationsverfahren kann die Übersichtlichkeit eines Node-Link- Diagramms gewährleistet werden. Dabei erkundet der Benutzer das Netzwerk schrittweise und kann so die Komplexität der Visualisierung steuern. Die statische, interaktive Exploration eines Graphen wird durch Filterung des globalen Layouts realisiert. Die dynamische, interaktive Exploration eines Graphen basiert dagegen auf einem modifizierten lokalen Layout. Mit jeder Änderung des darzustellenden Ausschnitts wird ein neues lokales Layout berechnet, welches dem vorherigen Layout ähnelt. Mit diesem Verfahren können lokale Zusammenhänge übersichtlich und nachvollziehbar visualisiert werden. Aber unterschiedliche Explorationsverläufe resultieren in unterschiedlichen Layouts, der Transfer des Wissens aus einer Exploration auf eine andere Exploration ist schwierig. Gibt es also eine Methode, die die Vorteile beider Verfahren verbinden kann? Eine Methode, die es dem Anwender erlaubt, sich auf seine Orientierung zu verlassen und trotzdem lokale Zusammenhänge des betrachteten Ausschnitts übersichtlich darzustellen? Dieser Frage widmet sich die vorliegende Arbeit. Es wird ein Ansatz zur Verschmelzung von Fotografien auf Node-Link-Diagramme übertragen, so dass 2 lokale Layouts zueinander ausgerichtet und miteinander verschmolzen werden können. Im Anschluss wird dieses Verfahren dann so erweitert, dass auch beliebige Mengen von lokalen Layouts miteinander verschmolzen werden können.

Show publication details

Bernard, Jürgen; Wilhelm, Nils; Scherer, Maximilian; May, Thorsten; Schreck, Tobias

TimeSeriesPaths: Projection-Based Explorative Analysis of Multivariate Time Series Data

2012

Skala, Vaclav (Ed.): Journal of WSCG Vol. 20 No. 1-3, 2012. Proceedings. Plzen: University of West Bohemia, 2012, pp. 97-106

International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision (WSCG) <20, 2012, Plzen, Czech Republic>

The analysis of time-dependent data is an important problem in many application domains, and interactive visualization of time-series data can help in understanding patterns in large time series data. Many effective approaches already exist for visual analysis of univariate time series supporting tasks such as assessment of data quality, detection of outliers, or identification of periodically or frequently occurring patterns. However, much fewer approaches exist which support multivariate time series. The existence of multiple values per time stamp makes the analysis task per se harder, and existing visualization techniques often do not scale well. We introduce an approach for visual analysis of large multivariate time-dependent data, based on the idea of projecting multivariate measurements to a 2D display, visualizing the time dimension by trajectories. We use visual data aggregation metaphors based on grouping of similar data elements to scale with multivariate time series. Aggregation procedures can either be based on statistical properties of the data or on data clustering routines. Appropriately defined user controls allow to navigate and explore the data and interactively steer the parameters of the data aggregation to enhance data analysis. We present an implementation of our approach and apply it on a comprehensive data set from the field of earth observation, demonstrating the applicability and usefulness of our approach.

Show publication details

May, Thorsten; Steiger, Martin; Davey, James; Kohlhammer, Jörn

Using Signposts for Navigation in Large Graphs

2012

Computer Graphics Forum, Vol.31 (2012), 3, pp. 985-994

Eurographics Conference on Visualization (EuroVis) <14, 2012, Vienna, Austria>

In this paper we present a new Focus & Context technique for the exploration of large, abstract graphs. Most Focus & Context techniques present context in a visual way. In contrast, our technique uses a symbolic representation: while the focus is a set of visible nodes, labelled signposts provide cues for the context - off-screen regions of the graph - and indicate the direction of the shortest path linking the visible nodes to these regions. We show how the regions are defined and how they are selected dynamically, depending on the visible nodes. To define the set of visible nodes we use an approach developed by van Ham and Perer that dynamically extracts a subgraph based on an initial focal node and a degree-of-interest function. This approach is extended to support multiple focal nodes. With the symbolic visualization, potentially interesting regions of a graph may be represented with a very small visual footprint. We conclude the paper with an initial user study to evaluate the effectiveness of the signposts for navigation tasks.

Show publication details

Bernard, Jürgen; Ruppert, Tobias; Goroll, Oliver; May, Thorsten; Kohlhammer, Jörn

Visual-Interactive Preprocessing of Time Series Data

2012

Kerren, Andreas (Ed.) et al.: Proceedings of SIGRAD 2012 : Interactive Visual Analysis of Data. Linköping: Linköping University Electronic Press, 2012. (Linköping Electronic Conference Proceedings 81), pp. 39-48

SIGRAD Conference <11, 2012, Växjö, Sweden>

Time series data is an important data type in many different application scenarios. Consequently, there are a great variety of approaches for analyzing time series data. Within these approaches different strategies for cleaning, segmenting, representing, normalizing, comparing, and aggregating time series data can be found. When combining these operations, the time series analysis preprocessing workflow has many degrees of freedom. To define an appropriate preprocessing pipeline, the knowledge of experts coming from the application domain has to be included into the design process. Unfortunately, these experts often cannot estimate the effects of the chosen preprocessing algorithms and their parameterizations on the time series. We introduce a system for the visual-interactive exploitation of the preprocessing parameter space. In contrast to 'black box'-driven approaches designed by computer scientists based on the requirements of domain experts, our system allows these experts to visual-interactively compose time series preprocessing pipelines by themselves. Visual support is provided to choose the right order and parameterization of the preprocessing steps. We demonstrate the usability of our approach with a case study from the digital library domain, in which time-oriented scientific research data has to be preprocessed to realize a visual search and analysis application.

Show publication details

May, Thorsten; Bannach, Andreas; Davey, James; Ruppert, Tobias; Kohlhammer, Jörn

Guiding Feature Subset Selection with an Interactive Visualization

2011

Miksch, Silvia (Ed.) et al.: IEEE Conference on Visual Analytics Science and Technology 2011. Proceedings : VAST 2011. New York: IEEE Press, 2011, pp. 111-120

IEEE Symposium on Visual Analytics Science and Technology (VAST) <6, 2011, Providence, RI, USA>

We propose a method for the semi-automated refinement of the results of feature subset selection algorithms. Feature subset selection is a preliminary step in data analysis which identifies the most useful subset of features (columns) in a data table. So-called filter techniques use statistical ranking measures for the correlation of features. Usually a measure is applied to all entities (rows) of a data table. However, the differing contributions of subsets of data entities are masked by statistical aggregation. Feature and entity subset selection are, thus, highly interdependent. Due to the difficulty in visualizing a high-dimensional data table, most feature subset selection algorithms are applied as a black box at the outset of an analysis. Our visualization technique, SmartStripes, allows users to step into the feature subset selection process. It enables the investigation of dependencies and interdependencies between different feature and entity subsets. A user may even choose to control the iterations manually, taking into account the ranking measures, the contributions of different entity subsets, as well as the semantics of the features.

Show publication details

May, Thorsten; Davey, James; Ruppert, Tobias

SmartStripes - Looking under the Hood of Feature Subset Selection Methods

2011

Miksch, Silvia (Ed.) et al.: EuroVA 2011 : International Workshop on Visual Analytics. Goslar: Eurographics Association, 2011, pp. 13-16

International Workshop on Visual Analytics (EuroVA) <2, 2011, Bergen, Norway>

We propose a visualization method for the diagnosis and interactive refinement of automatic techniques for feature subset selection. So-called filter techniques use statistical ranking measures to identify the most useful combination of features for further analysis. Usually a measure is applied to all entities of a data-table. The influence of atypical entities can distort the result, but this distortion may be masked by the statistical aggregation. Clearly, feature and entity subset selection are highly interdependent. Our technique, SmartStripes, intends to make this interdependency visible.

Show publication details

May, Thorsten; Davey, James; Kohlhammer, Jörn

Combining Details of the Chi-Square Goodness-of-Fit Test with Multivariate Data Visualization

2010

Kohlhammer, Jörn (Ed.) et al.: EuroVAST 2010 : International Symposium on Visual Analytics Science and Technology. Goslar: Eurographics Association, 2010, pp. 45-50

International Symposium on Visual Analytics Science and Technology (EuroVAST) <1, 2010, Bordeaux, France>

In this work, we combine KVMaps, a visualization technique presented in [May07] for the visualization of statistical aggregations in multivariate contingency tables, with the measures used for the statistical Chi-Square goodness-of-fit test. Goodness-of-fit tests are used to check whether a given distribution of values matches an expected distribution. A single test statistic is calculated to represent the deviation of the complete dataset. By visualizing the deviations for all entries in the contingency table, it is possible to identify the patterns in the distribution of data items, which contribute most to the overall deviation of the dataset. We present two use cases to illustrate how the information about the patterns can be used.

Show publication details

May, Thorsten; Davey, James

Combining Statistical Independence Testing, Visual Attribute Selection and Automated Analysis to Find Relevant Attributes for Classification

2010

MacEachren, Alan M. (Ed.) et al.: IEEE Conference on Visual Analytics Science and Technology 2010. Proceedings : VAST 2010. New York: IEEE Press, 2010, pp. 239-240

IEEE Symposium on Visual Analytics Science and Technology (VAST) <5, 2010, Salt Lake City, UT, USA>

We present an iterative strategy for finding a relevant subset of attributes for the purpose of classification in high-dimensional, heterogeneous data sets. The attribute subset is used for the construction of a classifier function. In order to cope with the challenge of scalability, the analysis is split into an overview of all attributes and a detailed analysis of small groups of attributes. The overview provides generic information on statistical dependencies between attributes. With this information the user can select groups of attributes and an analytical method for their detailed analysis. The detailed analysis involves the identification of redundant attributes (via classification or regression) and the creation of summarizing attributes (via clustering or dimension reduction). Our strategy does not prescribe specific analytical methods. Instead, we recursively combine the results of different methods to find or generate a subset of attributes to use for classification.

Show publication details

Peschek, Jens; May, Thorsten (Betreuer)

Erweiterung des Multi Attribute Data Mining zur Prozessanalyse

2010

Darmstadt, TU, Diplomarbeit, 2010

Die vorliegende Arbeit beschäftigt sich mit der Analyse von Projektdaten, die im Rahmen des Projektmanagements generiert wurden. Ziel dabei ist es, systematische Problemstellen zu identifizieren und daraus für zukünftige Projekte zu lernen und Konsequenzen ziehen zu können. Der verwendete Ansatz ist ein Data Mining-Verfahren, genauer gesagt ein Item Set Mining in Kombination mit einem Association Rule Mining. Um den Besonderheiten der vorliegenden Problematik gerecht werden zu können, müssen diese zu einem Multi Attribute Mining ausgedehnt werden. Dieses wird hier um die Verwendung von allgemeinen Relationen zur Gruppierung über diese Attribute erweitert. Auch eine Implementierung des neuen Ansatzes wird vorgestellt.

Show publication details

Keim, Daniel A.; Kohlhammer, Jörn; Mansmann, Florian; May, Thorsten; Wanner, Franz

Introduction

2010

Keim, Daniel A. (Ed.) et al.: Mastering the Information Age. Solving Problems with Visual Analytics. Goslar: Eurographics Association, 2010, pp. 1-6

Raw data has no value in itself, only the extracted information has value. Time and money are wasted and opportunities are lost. Success depends on availability of the right information. Visual analytics aims at making data and information processing transparent. Visual analytics combines the strengths of humans and computers.

Show publication details

Keim, Daniel A.; Kohlhammer, Jörn; Mansmann, Florian; May, Thorsten; Wanner, Franz

Visual Analytics

2010

Keim, Daniel A. (Ed.) et al.: Mastering the Information Age. Solving Problems with Visual Analytics. Goslar: Eurographics Association, 2010, pp. 7-18

Visual analytics is not easy to define, due to its multi-disciplinary nature involving multiple processes and the wide variety of application areas. An early definition was "the science of analytical reasoning facilitated by interactive human-machine interfaces". However, based on current practice, a more specific definition would be: "Visual analytics combines automated Visual analytics combines automated analysis with interactive visualisations analysis techniques with interactive visualisations for an effctive understanding, reasoning and decision making on the basis of very large and complex datasets". So, in terms of the goal of visual analytics, we can elaborate on this definition to state that visual analytics is the creation of tools and techniques to enable people to: - Synthesise information and derive insight from massive, dynamic, ambiguous, and often conflicting data. - Detect the expected and discover the unexpected. - Provide timely, defensible, and understandable assessments. - Communicate these assessment effectively for action

Show publication details

Kohlhammer, Jörn; May, Thorsten; Davey, James; Ruppert, Tobias

Visual Analytics - Verbindung von Analyseverfahren und Visualisierungstechniken

2010

IM - Die Fachzeitschrift für Information Management & Consulting, Vol.25 (2010), 3, pp. 10-17

Visual Analytics bietet eine Möglichkeit, um große Datenmengen besser zu verstehen. Wissen kann damit aus den gesammelten Unternehmensdaten gewonnen werden. Dieses Wissen kann für Unternehmen einen Wettbewerbsvorteil bedeuten. Bisher werden Unternehmenskennzahlen lediglich mithilfe von statistischen Modellen berechnet. Doch der Bezug zwischen den Daten, den Modellen und den Unternehmenszielen bleibt häufig unklar. Wenn die Visualisierung als wesentlicher Bestandteil des Analyseprozesses eingebunden wird, wird das Verständnis von den Modellen sowie auch von den Daten erhöht. Fehler in den zugrundeliegenden Annahmen können früher behoben werden und neu entdeckte Zusammenhänge können zu sicheren, an Fakten und Zielen überprüfbaren Kennzahlen beim Reporting führen.

Show publication details

Ruppert, Tobias; May, Thorsten; Kohlhammer, Jörn; Schreck, Tobias

Visuelle Analysen des Datensatzes: Wie versteckte Zusammenhänge sichtbar werden

2010

Trepte, Sabine (Ed.) et al.: Allgemeinbildung in Deutschland : Erkenntnisse aus dem SPIEGEL-Studentenpisa-Test. Wiesbaden: VS Verlag, 2010, pp. 87-104

Wenn Wissenschaftler Daten (z.B. des Studentenpisa-Tests des SPIEGEL) analysieren, stellen sie gemeinhin Hypothesen auf und überprüfen diese dann. In diesem Beitrag wird ein anderes Verfahren vorgestellt. Es handelt sich um ein exploratives Vorgehen, das es erlaubt, versteckte Zusammenhänge in großen und komplexen Datensammlungen zu finden. Dazu werden die Daten ohne die Formulierung einer bestimmten Fragestellung mittels interaktiver graphischer Darstellungen untersucht. Der Beitrag erläutert diese Herangehensweise und stellt drei Techniken vor, die für diesen Zweck am Fraunhofer-Institut für Graphische Datenverarbeitung in Darmstadt und an der TU Darmstadt entwickelt worden sind. Die Vielzahl von möglichen Aussagen über einen unbekannten Datensatz wird mit diesen und vergleichbaren Techniken auf jene reduziert, die es wert sind, genauer untersucht zu werden.

Show publication details

Sari, Fatih; May, Thorsten (Betreuer)

Klassifikation und Bewertung von Evaluierungsverfahren im Bereich der Informationsvisualisierung für die interaktive Datenanalyse

2009

Darmstadt, TU, Diplomarbeit, 2009

Das Thema der vorliegenden Arbeit ist die Klassifikation und Bewertung von Evaluierungsverfahren im Bereich der Informationsvisualisierung. Die Informationsvisualisierung als Forschungsgebiet gewinnt mit dem immensen Zuwachs an Informationen durch neuartige Technologien in der Informationstechnologie immer mehr an Bedeutung. Innerhalb der Gemeinschaft der Forscher wird unter anderem die Frage diskutiert, inwiefern sich die Techniken im vergangenen Jahrzehnt immer mehr von der praktischen Anwendung entfernt haben. Als eine Ursache hierfür wird der Mangel an Evaluierungstechniken und Methoden angeführt. Für die Evaluierung von Informationsvisualisierungsverfahren stehen verschiedene Evaluationstechniken und -methoden zur Verfügung. Die Erörterung dieser Techniken und Methoden mit ihren Vor- und Nachteilen sowie die praktische Umsetzung eines Benutzerexperiments mit Hierarchievisualisierungen ist das Ziel dieser Diplomarbeit.

Show publication details

Kohlhammer, Jörn; May, Thorsten; Hoffmann, Marcus

Visual Analytics for the Strategic Decision Making Process

2009

Amicis, Raffaele de (Ed.) et al.: GeoSpatial Visual Analytics : Geographical Information Processing and Visual Analytics for Environmental Security. Springer Science+Business Media, 2009. (NATO Science for Peace and Security Series C: Environmental Security), pp. 299-310

NATO Advanced Research Workshop on Geographical Information Processing and Visual Analytics for Environmental Security <2009, Trento, Italy>

The collection and storage of huge amounts of data is no longer a challenge by itself. However, rapidly growing data repositories are creating considerable challenges in many application areas. Visualizations that worked well with a few data items now produce confusing or illegible displays. Decision-makers struggle to act based on a severely restricted understanding of the situation. The goal of Visual Analytics is to overcome this information overload and create new opportunities with these large amounts of data and information. The key challenge is to intelligently combine visualization techniques and analytic algorithms, and to enable the human expert to guide the decision making process. This chapter covers interesting and relevant previous work on situation awareness, naturalistic decision making, and decision-centred visualization. These concepts are put into the context of Visual Analytics research and are further illustrated by application examples.

Show publication details

Bienek, Markus; May, Thorsten (Betreuer)

Graphen-Layout-Algorithmen für Graph-Visualisierung - Entwicklung einer generellen Struktur auf Basis von bestehenden Algorithmen

2008

Darmstadt, Hochschule, Bachelor Thesis, 2008

Die Analyse von Graphen und Netzwerken ist ein wichtiges Thema der diskreten Mathematik mit einer stetig wachsenden Anzahl von Anwendungsgebieten. Im Bereich Visual Analytics tauchen Graphen sowohl in der Darstellung von Daten (soziale Netzwerke, wirtschaftliche Beziehungen, Dokumentverbindungen, etc.), aber auch in der visuellen Darstellung abstrakter analytischer Modelle auf, wie etwa Entscheidungsbäume, probabilistische Netzwerke, etc. Technisch macht es aber keinen Unterschied, welche Graphen mit den Visualisierungstechniken angezeigt werden. Da die Wahl einer geeigneten Technik, aber durchaus von der konkreten Aufgabenstellung und der Art des Graphen abhängt, soll die Anpassung der Visualisierung an diese Aspekte in Zukunft möglichst effizient gestaltet werden. In der Vergangenheit war die Entwicklung der Visualisierungstechniken für Graphen und Hierarchien eher heterogen. Unterschiedliche Algorithmen für das Layout und die Abbildung visueller Strukturen wurden in unterschiedlichen Programmen und Zielsetzungen umgesetzt. In der vorliegenden Arbeit wurden diese Entwicklungen untersucht, und Software Design-Patterns identifiziert, die schließlich für die Entwicklung von einer "Standardvisualisierungtechnik" für Graphen notwendig sind. Besonders notwendig waren dabei die klare strukturelle Trennung der Algorithmen in Komponenten für das Zeichnen von Kanten und Knoten, das Layout des Graphen, die jeweils unabhängig voneinander ausgetauscht werden können. Besonderes Ziel dabei war auch, dass die zusätzliche Flexibilität keine spürbaren Einbußen an Performance mit sich bringt. Die Ergebnisse der Arbeit wurden in einem Prototypen, der alte Techniken in dem neuen Standard konsolidiert umgesetzt. Die Funktionalität dieses Prototypen wurde durch die unabhängige Rekombinierbarkeit verschiedener wesentlich erweitert.

Show publication details

May, Thorsten; Kohlhammer, Jörn

Towards Closing the Analysis Gap: Visual Generation of Decision Supporting Schemes from Raw Data

2008

Computer Graphics Forum, Vol.27 (2008), 3, pp. 911-918

Eurographics and IEEE VGTC Symposium on Visualization (EuroVis) <10, 2008, Eindhoven, The Netherlands>

The derivation, manipulation and verification of analytical models from raw data is a process which requires a transformation of information across different levels of abstraction. We introduce a concept for the coupling of data classification and interactive visualization in order to make this transformation visible and steerable for the human user. Data classification techniques generate mappings that formally group data items into categories. Interactive visualization includes the user into an iterative refinement process. The user identifies and selects interesting patterns to define these categories. The following step is the transformation of a visible pattern into the formal definition of a classifier. In the last step the classifier is transformed back into a pattern that is blended with the original data in the same visual display. Our approach allows in intuitive assessment of a formal classifier and its model, the detection of outliers and the handling of noisy data using visual pattern-matching. We instantiated the concept using decision trees for classification and KVMaps as the visualization technique. The generation of a classifier from visual patterns and its verification is transformed from a cognitive to a mostly pre-cognitive task.

Show publication details

May, Thorsten; Kohlhammer, Jörn

Visual Verification of Hypotheses

2008

Bebis, George (Ed.) et al.: Advances in Visual Computing. Proceedings Part II : ISVC 2008. Berlin, Heidelberg, New York: Springer, 2008. (Lecture Notes in Computer Science (LNCS) 5359), pp. 31-42

International Symposium on Visual Computing (ISVC) <4, 2008, Las Vegas, NV, USA>

The analytical derivation of a hypothesis is a process, that requires a transformation of information between raw data and an analytical model. Even though much effort has been spent to support the creation of hypotheses both by algorithmic and visual means, much less research has been done on how the process can be reversed for the verification of existing hypotheses. An evaluation of empirical hypotheses must be grounded in raw data and may require many tedious drill-downs, especially for complex data. We present a concept combining an analytical technique for the representation of hypotheses and their transformation into the data-space. We also show visualization techniques for the formalization of the hypothesis in the analytical space and its visual evaluation in the data space. The evaluation is supported by a visual-matchmaking between original raw data and a modification of this data based upon the assumptions implied by the hypothesis.

Show publication details

May, Thorsten

Working with Patterns in Large Multivariate Datasets - Karnaugh-Veitch-Maps Revisited

2007

Banissi, Ebad (Ed.) et al.: 11th International Conference on Information Visualization. Proceedings : IV 2007. Los Alamitos, Calif.: IEEE Computer Society, 2007, pp. 277-284

IEEE International Conference on Information Visualization (IV) <11, 2007, Zurich, Switzerland>

We present an interactive visualization method for the multivariate analysis of large and complex datasets, based on the layout of Karnaugh-Veitch-diagrams. Working on data categories, we additionally provide an interactive partitioning of value ranges of ordinal types. Multivariate dependencies manifest on the map in characteristic color patterns. These patterns are representations of subsets of the data or attributes which embody significant information. While the task of the user is to identify visual patterns of particular interest to him by clicking on the map, the software identifies a minimal representation of the corresponding subset and gives a visual feedback. Hence, user and machine cooperate on the basis of a strong visual coupling in short iterative cycles. During this process the trade-off between the accuracy and simplicity of a representation, which is crucial for any type of the building of analytical models, can be found in an effective way.

Show publication details

Chen, Kefeng; May, Thorsten (Betreuer)

Eine wieder verwendbare dynamische Ressourcenverteilung für verschiedene Tasks auf einem Workstationcluster

2006

Darmstadt, TU, Diplomarbeit, 2006

The parallel computing is a very important approach to improve the efficiency of certain programs. Parts of a parallel program run simultaneously on several processors or computers so that the computing time will be greatly reduced. Computer clusters are several computers linked with Ethernet . It's the cheapest way to utilize the power of parallel computing and can be logically infinitely extended. To take advantage of a cluster, one must employ certain APIs like MPI or PVM. Such APIs use process as the smallest calculating unit to distribute on the cluster. With these APIs one must explicit program the communication between processes. This diplom thesis proposed a new way to program the cluster: using the thread. A framework were designed and a prototype of that framework, with which a program can distribute it's threads on a cluster, was implemented based upon MPI. The framework takes over the communication between threads and provides tools such as mutex, semaphore und wait condition to control the work flow between threads. A default stepwise load balancing algorithm was also provided to distribute the workload on the cluster as balanced as possible. Design patterns were used to provide flexibility so that the algorithms can be easily updated. The prototype was tested at the end on a workstation cluster with the marching cubes algorithm.

Show publication details

Decavel, Quentin; May, Thorsten (Betreuer)

Entwurf und Realisierung einer interaktiven Visualisierung für die Konfiguration und Steuerung von Servernetzwerken

2006

Darmstadt, TU, Diplomarbeit, 2006

The goal of this work is to define and implement a module for "Visionapp Control Center". "Visionapp Control Center" is an application which aims to provide system administrators with a graphical user interface for the automatization of server management inside a network. The module provides an easy way to quickly create server networks, to display information about these networks, and to visualize them as completely as possible. A central part of this work is thus the development of a so-called "Fisheye"-view. Fisheye-views help visualize heterogenous amounts of data by allowing the user to focus on any subset of data while still keeping the whole data set in sight. In our case, the user can choose to focus on a particular connected set of objects and connections inside the network, while still being able to see the whole network. This part is essentially a mathematical one, with the characterization of network objects as elements that move on the surface of a sphere. Another very important aspect of this work is the user-friendliness of the module: the user must be able to create, visualize and expand networks as easily as possible. This demand leads to a need for appropriate user-event handling, to facilitate the user's task.

Show publication details

Keim, Daniel A.; Kohlhammer, Jörn; May, Thorsten; Thomas, Jim

Event Summary of the Workshop on Visual Analytics

2006

Computers & Graphics, Vol.30 (2006), 2, pp.284-286

The increasing power of computer, simulation and sensor technologies enables us to produce and store data at ever growing rates. Nevertheless, in virtually every application area, data has no value in itself: In very large amounts, data actually hides the important information and makes information more and more difficult to acquire. Decision makers and analysts are often confronted with vast amounts of complex, seemingly inconsistent or incomplete data, which are derived from a number of heterogeneous sources. The strategic effort of Fraunhofer IGD is pursued in collaboration with other national and international experts in this area. The recent Workshop on Visual Analytics was jointly organized with Prof. Keim, University of Konstanz and received endorsement by Jim Thomas, PNNL, who is head of the extensive Visual Analytics efforts in the United States.

Show publication details

Duchstein, Patrick; May, Thorsten (Betreuer)

Extraktion und Visualisierung von Relationen in Textdokumentensammlungen auf der Basis einer Wortfeldanalyse

2006

Darmstadt, TU, Diplomarbeit, 2006

In der Arbeit wird ein System zur Analyse und Visualisierung von Dokumentsammlungen auf Basis einer Wortfeldanalyse konzipiert und vorgestellt. Dazu wird eine Menge von Dokumenten aus einem jeweiligen Quellformat zunächst in Klartexte umgewandelt und Metadaten wie Autor und Titel werden extrahiert; Wörter werden auf die korrespondierenden Stammformen abgebildet. Die Dokumente werden dann in ein Vektorraummodell überführt, so dass für jedes Dokument ein Vektor existiert, welcher aus dessen gewichteten Termen besteht. Zur Dimensionalitätsreduktion dieser Vektoren und somit zur Wortfeldanalyse findet die Methodik des Latent Semantic Indexing (LSI) Anwendung, unter Berücksichtigung verschiedener Methoden zur Singulärwertzerlegung; weitere Methoden für eine Dimensionalitätsreduktion werden untersucht. Verschiedene Ähnlichkeitsmaße zwischen Dokumentpaaren werden vorgestellt und analysiert. Die Visualisierung erfolgt sowohl zwei- als auch dreidimensional mit Hilfe von Self-Organizing Maps (SOM). Die dreidimensionale Darstellung erfolgt unter Zuhilfenahme von Volumenrendering. Als weitere Visualisierungsmethode steht ein Plot der Hauptachsen im reduzierten Vektorraummodell zur Verfügung. Weiterhin können die Dokumente anhand ihrer Ähnlichkeit gruppiert und anhand der Partitionierung in den Visualisierungen farblich dargestellt und somit unterschieden werden. Die im Rahmen dieser Arbeit entstandenen Software soll weitere Verwendung in einem allgemeinen Visual Analytics Framework finden.

Show publication details

May, Thorsten

Conversion and Encoding of Simplical Progressive Grids of Arbitrary Dimension

2005

Schulze, Thomas (Ed.) et al.: Simulation und Visualisierung 2005. Proceedings. Erlangen: SCS-European Publishing House, 2005, pp. 233-246

Simulation und Visualisierung (SimVis) <16, 2005, Magdeburg>

We present an irregular simplical grid for the progressive display and transmission of CFD-simulation datasets of arbitrary dimension. Spatial and temporal dimensions are treated equally to exploit the coherencies between different timeframes of a simulation. The data is hierarchically arranged by its importance to minimize the amount of data to be stored or transmitted. We use a recursive partitioning scheme for the grid cells, which tests a number of candidate partitionings by their quality. Furthermore a progressive encoding scheme is proposed, which minimizes the effort to store the topology of the grid. Possible applications stretch beyond progressive transmission into areas of data compression and visualization based on multiple levels-of-detail.

Show publication details

May, Thorsten; Schneider, Sascha; Schmidt, Michael; Luckas, Volker

Fast Scalar- & Vectorfield Visualization Using a New Progressive Grid Class

2003

Schulze, Thomas (Ed.) et al.: Simulation und Visualisierung. Proceedings 2003. Ghent: SCS-Europe, 2003, pp. 89-101

Tagung Simulation und Visualisierung <2003, Magdeburg>

We present a new generic progressive data format and framework for CFD-data, which is specialized for scientific data simulation: The progressive grids. While using arbitrary types of cells and partitioning schemes, simulation data can be stored progressively. Hierachically arranged by its level of detail, the data can be adapted efficiently to many systems with different ressources. Maintaining a modular approach, we are able to implement, test and update different kinds of progressive grids by adjusting the corresponding parts of the system.

Show publication details

Schneider, Sascha; May, Thorsten; Schmidt, Michael

Parallel Architecture of an Interactive Scientific Visualization System for Large Datasets

2003

Reiners, Dirk (Ed.): OpenSG 2003 : Concepts and Components of Software Frameworks for Interactive 3D Graphics. Aire-la-Ville: The Eurographics Association, 2003, pp. 15-21

OpenSG Symposium <2, 2003, Darmstadt, Germany>

In this paper we describe a further development state of our system which is able to compute actual scientific and realistic visualization methods in parallel. This paper is related to the basic work we presented in one of our previous articles. Our system is capable to integrate easily in modern VR renderers like for example Open Inventor, Coin and OpenSG. Our approach is designed for processing large datasets which usually are the result of physically based simulation algorithms and programs. Using our techniques it is even more feasible to manage similar visualization problems for other large amounts of data (e.g. medicinal CT-scans or large geometries) in the context of displaying interactively.

Show publication details

Maus, Andreas; Groch, W.-D. (Betreuer); May, Thorsten (Betreuer)

Realitätsnahe Visualisierung von Feuer basierend auf CFD-Daten

2003

Darmstadt, FH, Diplomarbeit, 2003

Show publication details

Schneider, Sascha; May, Thorsten; Schmidt, Michael

Rendering Large (Volume) Datasets: A new Parallel Visualization System

2003

Skala, Vaclav (Ed.): Journal of WSCG Volume 11 No. 3. Proceedings. Plzen: University of West Bohemia, 2003, pp. 418-424

International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision (WSCG) <11, 2003, Plzen, Czech Republic>

In this paper we describe a basis for a system that is able to compute actual scientific and realistic visualization methods in parallel. It is capable to integrate easily in modern VR renderers like for example Open Inventor [SGIa], Coin [Coi] and OpenSG [Vos02]. Our approach is advantageous for processing large datasets which usually are the result of physically based simulation algorithms and programs. Using our techniques it is even more feasible to manage similar visualization problems for other large amounts of data (e.g. medicinal CT-scans or large geometries) in the context of displaying interactively.

Show publication details

May, Thorsten; Schneider, Sascha; Schmidt, Michael; Luckas, Volker

The Progressive Grid: Introducing a New Grid Class for Efficient CFD Visualization

2003

Banicescu, Ioana (Ed.): High Performance Computing. Proceedings. San Diego: Society for Modeling and Simulation International, 2003, pp 115-120

High Performance Computing Symposium (HPC) <2003, Orlando, Florida, USA>

We present a new generic progressive data format and framework for CFD-Data, which is specialized for scientific data simulation: The progressive grids. While using arbitrary types of cells and partitioning schemes, simulation data can be stored progressively. Hierarchically arranged by its level of detail the data can be adapted efficiently to many systems with different ressources. Maintaining a modular approach, we are able to implement, test and update different kinds of progressive grids by adjusting the corresponding parts of the system.

Show publication details

May, Thorsten; Schneider, Sascha; Luckas, Volker

Parallel Real Time Fluid Simulation and Animation with Fractal Optical Refinements

2002

Amborski, Krzysztof (Ed.) et al.: 16th European Simulation Multiconference: Modelling and Simulation 2002 : ESM 2002. Ghent: SCS-Europe, 2002, pp. 224-228

European Simulation Multiconference: Modelling and Simulation <16, 2002, Darmstadt>

In this paper we describe the parallelization of an unconditional stable solution scheme of the Navier Stokes equation which is suitable for animation purposes. Profiting from its unconditional stability we are able to use simulation time steps which are larger than the animation timesteps and interpolate between them. Furthermore we present additional improvements in the visualization quality of the simulated output by using fractal textures in the 3D texture renderer post process for optical refinement using the scalar and vector fields as input parameters Our technique facilitate real-time animation rates of several examples on current multi-processor workstations and PC's

Show publication details

Klein, Reinhard; May, Thorsten; Schneider, Sascha; Weber, Andreas

Real-Time Fluid Animation by Parallel and Stable Solution Techniques

2001

Schilling, Andreas (Ed.): Festschrift zum 60. Geburtstag von Wolfgang Straßer : WSI / Wilhelm Schickard-Institut für Informatik: Sonderheft. Tübingen: Wilhelm-Schickard-Institut für Informatik, 2001, pp. 91-107

We describe new techniques for the parallelization of an unconditional stable solution scheme of the Navier-Stokes equation that has recently been introduced for animation purposes. The parallel improvement over existing schemes eleminates the sequential bottleneck. Aditionally our solution method makes use of time steps which are bigger than the framerate for interpolation between them. We achieve real-time animation rates foe several examples on current multi-processor workstations and PC's.

Show publication details

Bryborn, Matthias; Klein, Reinhard; May, Thorsten; Schneider, Sascha; Weber, Andreas

A Portable, Parallel, Real-Time Animation-System for Turbulent Fluids

2000

Guizani, M. (Ed.) et al.: Parallel and Distributed Computing and Systems. Proceedings of the IASTED International Conference. Anaheim: IASTED/ACTA Press, 2000, pp. 394-400

IASTED International Conference on Parallel and Distributed Computing Systems <2000, Las Vegas, NV, USA>

We describe our parallelization of an unconditionally stable solution scheme of the Navier-Stokes equation that has recentlybeen used for animation purposes. Our parallelization gives good speed-ups on current multi-processor workstations. These speed-ups close the gap that existed towards real time animation for several 3D-examples. In addition to parallelizing the solver software the integration of a fast 3D-volume renderer into the parallel framework has been an important step to achieve this goal. Our parallelization is portable on workstations and PC s running under the UNIX or the WIN32 operating systems.