IVAPP 2014 Abstracts


Area 1 - Abstract Data Visualization

Full Papers
Paper Nr: 1
Title:

Software Feathers - Figurative Visualization of Software Metrics

Authors:

Fabian Beck

Abstract: In order to give the code entities of a software system a discernible and recognizable face, this paper presents Software Feathers, an approach for mapping characteristics of object-oriented code entities to features of artificially generated feathers. A parameterized drawing algorithm is described that generates pseudo-realistic feathers as 2D graphics. The parameters of the feathers are connected to characteristic software metrics measuring, among others, the size, complexity, and quality of interfaces and classes. Applying the approach demonstrates that categories of code entities can be discerned, problems in the code can be detected, and the evolution of code can be studied. A promising application is embedding the feathers into documentation and IDEs for improving navigation and unobtrusively educating software developers in software metrics.

Paper Nr: 6
Title:

Generalized Pythagoras Trees for Visualizing Hierarchies

Authors:

Fabian Beck, Michael Burch, Tanja Munz, Lorenzo Di Silvestro and Daniel Weiskopf

Abstract: Pythagoras Trees are fractals that can be used to depict binary hierarchies. But this binary encoding is an obstacle for visualizing hierarchical data such as file systems or phylogenetic trees, which branch into n subhierarchies. Although any hierarchy can be modeled as a binary one by subsequently dividing n-ary branches into a sequence of n - 1 binary branches, we follow a different strategy. In our approach extending Pythagoras Trees to arbitrarily branching trees, we only need a single visual element for an n-ary branch instead of spreading the binary branches along a strand. Each vertex in the hierarchy is visualized as a rectangle sized according to a metric. We analyze several visual parameters such as length, width, order, and color of the nodes against the use of different metrics. The usefulness of our technique is illustrated by two case studies visualizing directory structures and a large phylogenetic tree. We compare our approach with existing tree diagrams and discuss questions of geometry, perception, readability, and aesthetics.

Paper Nr: 17
Title:

An Inverse Distance-based Potential Field Function for Overlapping Point Set Visualization

Authors:

Jevgenijs Vihrovs, Krišjānis Prūsis, Kārlis Freivalds, Pēteris Ručevskis and Valdis Krebs

Abstract: In this paper we address the problem of visualizing overlapping sets of points with a fixed positioning in a comprehensible way. A standard visualization technique is to enclose the point sets in isocontours generated by bounding a potential field function. The most commonly used functions are various approximations of the Gaussian distribution. Such an approach produces smooth and appealing shapes, however it may produce an incorrect point nesting in generated regions, e.g. some point is contained inside a foreign set region. We introduce a different potential field function that keeps the desired properties of Gaussian distribution, and in addition guarantees that every point belongs to all its sets’ regions and no others, and that regions of two sets with no common points have no overlaps. The presented function works well if the sets intersect each other, a situation that often arises in social network graphs, producing regions that reveal the structure of their clustering.

Paper Nr: 19
Title:

Template-based Treemaps to Preserve Spatial Constraints

Authors:

Natallia Kokash, Bernard de Bono and Joost Kok

Abstract: Treemapping is a method for displaying hierarchical data using nested rectangles. Each branch of the tree is given its rectangle, which then is tiled with smaller rectangles representing sub-branches. A node’s shape has an area proportional to a specified dimension of the data. To create a treemap, one must define a tiling algorithm, i.e., a way to divide a rectangle into sub-rectangles of specified areas. There are several problems with existing tiling algorithms: (i) they allow tiles to shift when the main window or some of the tiles are resized affecting the user perception of the information; (ii) they do not allow users to place selected elements into desired positions w.r.t. each other. In this paper, we present a method for creating treemaps with customized layouts. The method is based on reusable templates and supports zooming into specific areas of the treemap without affecting its initial layout. We illustrate the use of template-based treemaps to the visualization of biomedical data. Furthermore, we present an algorithm for automated generation of layouts satisfying positional constraints for a particular class of constraints.

Paper Nr: 20
Title:

Visualization of Varying Hierarchies by Stable Layout of Voronoi Treemaps

Authors:

Sebastian Hahn, Jonas Trümper, Dominik Moritz and Jürgen Döllner

Abstract: Space-restricted techniques for visualizing hierarchies generally achieve high scalability and readability (e.g., tree maps, bundle views, sunburst). However, the visualization layout directly depends on the hierarchy, that is, small changes to the hierarchy can cause wide-ranging changes to the layout. For this reason, it is difficult to use these techniques to compare similar variants of a hierarchy because users are confronted with layouts that do not expose the expected similarity. Voronoi treemaps appear to be promising candidates to overcome this limitation. However, existing Voronoi treemap algorithms do not provide deterministic layouts or assume a fixed hierarchy. In this paper we present an extended layout algorithm for Voronoi treemaps that provides a high degree of layout similiarity for varying hierarchies, such as software-system hierarchies. The implementation uses a deterministic initial-distribution approach that reduces the variation in node positioning even if changes in the underlying hierarchy data occur. Compared to existing layout algorithms, our algorithm achieves lower error rates with respect to node areas in the case of weighted Voronoi diagrams, which we show in a comparative study.

Paper Nr: 27
Title:

Visualizations for Text Re-use

Authors:

Stefan Jänicke, Annette Geßner, Marco Büchler and Gerik Scheuermann

Abstract: In this paper, we present various visualizations for the Text Re-use found between texts of a collection to support humanists in answering a broad palette of research questions. When juxtaposing all texts of a corpus in the form of tuples, we propose the Text Re-use Grid as a distant reading method that emphasizes text tuples with systematic or repetitive Text Re-use. In contrast, the Text Re-use Browser allows for close reading of the Text Re-use between the two texts of a tuple. Additionally, we present Sentence Alignment Flows to improve the readability for Text Variant Graphs on sentence level that are used to compare various text editions to each other. Finally, we portray findings of the humanists of our project using the proposed visualizations.

Paper Nr: 34
Title:

Visualization of Remote Sensing Imagery by Sequential Dimensionality Reduction on Graphics Processing Unit

Authors:

Safa A. Najim and Ik Soo Lim

Abstract: This paper introduces a new technique called Sequential Dimensionality Reduction (SDR), to visualize remote sensing imagery. The DR methods are introduced to project directly the high dimensional dataset into a low dimension space. Although they work very well when original dimensions are small, their visualizations are not efficient enough with large input dimensions. Unlike DR, SDR redefines the problem of DR as a sequence of multiple dimensionality reduction problems, each of which reduces the dimensionality by a small amount. The SDR can be considered as a generalized idea which can be applied to any method, and the stochastic proximity embedding (SPE) method is chosen in this paper because its speed and efficiency compared to other methods. The superiority of SDR over DR is demonstrated experimentally. Moreover, as most DR methods also employ DR ideas in their projection, the performance of SDR and 20 DR methods are compared, and the superiority of the proposed method in both correlation and stress is shown. Graphics processing unit (GPU) is the best way to speed up the SDR method, where the speed of execution has been increased by 74 times in comparison to when it was run on CPU.

Paper Nr: 35
Title:

Data Visualization using Decision Trees and Clustering

Authors:

Olivier Parisot, Yoanne Didry, Pierrick Bruneau and Benoît Otjacques

Abstract: Decision trees are simple and powerful tools for knowledge extraction and visual analysis. However, when applied to complex datasets available nowadays, they tend to be large and uneasy to visualize. This difficulty can be overcome by clustering the dataset and representing the decision tree of each cluster independently. In order to apply the clustering more efficiently, we propose a method for adapting clustering results with a view to simplifying the decision tree obtained from each cluster. A prototype has been implemented, and the benefits of the proposed method are shown using the results of several experiments performed on the UCI benchmark datasets.

Short Papers
Paper Nr: 3
Title:

Visualizing Hierarchy Changes by Dynamic Indented Plots

Authors:

Michael Burch, Tanja Blascheck, Christine Louka and Daniel Weiskopf

Abstract: Visualizing static hierarchical organizations has been a focus of information visualization for many years, but it still remains challenging to produce visual representations of evolving, i.e. dynamic hierarchies. In this paper we extend the concept of indented plots to also support the depiction of changing hierarchies. We exploit the concept of static diagrams in order to support the preservation of a viewer’s mental map. Changes between subsequent hierarchies are precomputed and visually indicated by color coded straight or curved lines depending on the type of change. Interactive features can be used to aggregate or collapse the dynamic hierarchy data in the supported dimensions, i.e. the vertex as well as the time dimension on different levels of hierarchical and temporal granularity. The usefulness of our technique is illustrated by means of a bibliography dataset where we show the changes in the yearly prefix tree acquired by extracting words occuring most frequently in more than 2,000,000 paper titles from the digital library DBLP.

Paper Nr: 11
Title:

Storyboard Augmentation of Process Model Grammars for Stakeholder Communication

Authors:

Nardella Kathleen, Brown Ross and Simone Kriglstein

Abstract: Process models are often used to visualize and communicate workflows to involved stakeholders. Unfortunately, process modeling notations can be complex and need specific knowledge to be understood. Storyboards, as a visual language to illustrate workflows as sequences of images, provide natural visualization features that allow for better communication, to provide insight to people from non-process modeling expert domains. This paper proposes a visualization approach using a 3D virtual world environment to visualize storyboards for business process models. A prototype was built to present its applicability via generating output with examples of five major process model patterns and two non-trivial use cases. Illustrative results for the approach show the promise of using a 3D virtual world to visualize complex process models in an unambiguous and intuitive manner.

Paper Nr: 22
Title:

Multi-level Visualisation using Gaussian Process Latent Variable Models

Authors:

Shahzad Mumtaz, Darren R. Flower and Ian T. Nabney

Abstract: Projection of a high-dimensional dataset onto a two-dimensional space is a useful tool to visualise structures and relationships in the dataset. However, a single two-dimensional visualisation may not display all the intrinsic structure. Therefore, hierarchical/multi-level visualisation methods have been used to extract more detailed understanding of the data. Here we propose a multi-level Gaussian process latent variable model (MLGPLVM). MLGPLVM works by segmenting data (with e.g. K-means, Gaussian mixture model or interactive clustering) in the visualisation space and then fitting a visualisation model to each subset. To measure the quality of multi-level visualisation (with respect to parent and child models), metrics such as trustworthiness, continuity, mean relative rank errors, visualisation distance distortion and the negative log-likelihood per point are used. We evaluate the MLGPLVM approach on the ‘Oil Flow’ dataset and a dataset of protein electrostatic potentials for the ‘Major Histocompatibility Complex (MHC) class I’ of humans. In both cases, visual observation and the quantitative quality measures have shown better visualisation at lower levels.

Paper Nr: 25
Title:

A Computational Metric of the Quality of Circulation in Interior Spaces

Authors:

Arash Bahrehmand, Alun Evans and Josep Blat

Abstract: Space, in terms of interior and exterior design, is one of the most important issues facing all architects. In particular the movement of people through sequences of spaces forms a large part of the circulation problem in architecture planning. Although several studies have applied network models on urban analysis to take advantage of graph based queries, understanding interior design principles based on graph attributes shows potential for further research. This paper presents a computational solution to analyse, visualize, and evaluate the circulation quality of indoor spaces. To achieve it, first we create a grid graph based on a geometrical representation of space. Using this grid, a semantic weighted graph is generated, that helps us to provide a measured score for the circulation of people in a given space. The results were tested against architects’ scoring, showing that the measure is adequate. We also discuss the efficiency of our approach.

Paper Nr: 33
Title:

A Fraud Detection Visualization System Utilizing Radial Drawings and Heat-maps

Authors:

Evmorfia N. Argyriou, Antonios Symvonis and Vassilis Vassiliou

Abstract: We present a prototype system developed in cooperation with a business organization that combines information visualization and pattern-matching techniques to detect fraudulent activity by employees. The system is built upon common fraud patterns searched while trying to detect occupational fraud suggested by internal auditors of a business company. The main visualization of the system consists of a multi-layer radial drawing that represents the activity of the employees and clients. Each layer represents a different examined pattern whereas heat-maps indicating suspicious activity are incorporated in the visualization. The data are first preprocessed based on a decision tree generated by the examined patterns and each employee is assigned a value indicating whether or not there exist indications of fraud. The visualization is presented as an animation and the employees are visualized one by one according to their severity values together with their related clients.

Paper Nr: 42
Title:

Context-Specific Sentiment Lexicon Expansion via Minimal User Interaction

Authors:

Raheleh Makki, Stephen Brooks and Evangelos E. Milios

Abstract: One of the important factors in the performance of sentiment analysis methods is having a comprehensive sentiment lexicon. However, since sentiment words have different polarities not only in different domains, but also in different contexts within the same domain, constructing such context-specific sentiment lexicons is not an easy task. The high costs of manually constructing such lexicons motivate researchers to create automatic methods for finding sentiment words and assigning their polarities. However, existing methods may encounter ambiguous cases with contradictory evidence which are hard to automatically resolve. To address this problem, we aim to engage the user in the process of polarity assignment and improve the quality of the generated lexicon via minimal user effort. A novel visualization is employed to present the results of the automatic algorithm, i.e., the extracted sentiment pairs along with their polarities. User interactions are provided to facilitate the supervision process. The results of our user study demonstrate (1) involving the user in the polarity assignment process improves the quality of the generated lexicon significantly, and (2) participants in the study preferred our visual interface and conveyed that it is easier to use compared to a text-based interface.

Paper Nr: 45
Title:

Visual Exploration of Relationships between Document Clusters

Authors:

Ilir Jusufi, Andreas Kerren, Jiayi Liu and Björn Zimmer

Abstract: The visualization of networks with additional attributes attached to the network elements is one of the ongoing challenges in the information visualization domain. Such so-called multivariate networks regularly appear in various application fields, for instance, in data sets which describe friendship networks or co-authorship networks. Here, we focus on networks that are based on text documents, i.e., the network nodes represent documents and the edges show relationships between them. Those relationships can be derived from common topics or common co-authors. Attached attributes may be specific keywords (topics), keyword frequencies, etc. The analysis of such multivariate networks is challenging, because a deeper understanding of the data provided depends on effective visualization and interaction techniques that are able to bring all types of information together. In addition, automatic analysis methods should be used to support the analysis process of potentially large amounts of data. In this paper, we present a visualization approach that tackles those analysis problems. Our implementation provides a combination of new techniques that shows intra-cluster and inter-cluster relations while giving insight into the content of the cluster attributes. Hence, it facilitates the interactive exploration of the networks under consideration by showing the relationships between node clusters in context of network topology and multivariate attributes.

Paper Nr: 48
Title:

A Structured Approach for Conducting a Series of Controlled Experiments in Software Visualization

Authors:

Richard Müller, Pascal Kovacs, Jan Schilbach, Ulrich W. Eisenecker, Dirk Zeckzer and Gerik Scheuermann

Abstract: In the field of software visualization controlled experiments are an important instrument to investigate the specific reasons, why some software visualizations excel the expectations on providing insights and ease task solving while others fail doing so. Despite this, controlled experiments in software visualization are rare. A reason for this is the fact that performing such evaluations in general, and particularly performing them in a way that minimizes the threats to validity, is hard to accomplish. In this paper, we present a structured approach on how to conduct a series of controlled experiments in order to give empirical evidence for advantages and disadvantages of software visualizations in general and of 2D vs. 3D software visualizations in particular.

Paper Nr: 49
Title:

Level Set Trees with Enhanced Marginal Density Visualization

Authors:

Kyösti Karttunen, Lasse Holmström and Jussi Klemelä

Abstract: We study level set tree methods to analyze and visualize multivariate data. The probability density function of the underlying distribution is estimated using a kernel density estimator, and the density estimate is visualized using level set trees. These trees can be used to analyze the mode structure of a function. We show how level set trees can be used to enhance more traditional density function visualization tools, like marginal densities and slices of the density. The method is applied to flow cytometry data.

Paper Nr: 53
Title:

A Perceptive Insight into Cities Patterns by Visualizing Urban Economies

Authors:

Luca Piovano, Alberto Andréu, Iris Galloso and Claudio Feijóo

Abstract: Urban economic activities are an essential facet in defining city identity. Traditional approaches rely very often on the most theoretical and quantitative features of the studies, excluding de-facto a direct association between those findings and the tangible subject of the analysis. To fill the gap, the Big Data era and information visualization methodologies could help analysts, stakeholders and general audience to gain a new insight on the field. In this paper, we want to provide some food for thought about new opportunities arising in visual urban economies as well as present some visual results on possible scenarios.

Posters
Paper Nr: 4
Title:

Visual Analysis of Time-Dependent Multivariate Data from Dairy Farming Industry

Authors:

Lorenzo Di Silvestro, Michael Burch, Margherita Caccamo, Daniel Weiskopf, Fabian Beck and Giovanni Gallo

Abstract: This paper addresses the problem of analyzing data collected by the dairy industry with the aim of optimizing the cattle-breeding management and maximizing profit in the production of milk. The amount of multivariate data from daily records constantly increases due to the employment of modern systems in farm management, requiring a method to show trends and insights in data for a rapid analysis. We have designed a visual analytics system to analyze time-varying data. Well-known visualization techniques for multivariate data are used next to novel methods that show the intrinsic multiple timeline nature of these data as well as the linear and cyclic time behavior. Seasonal and monthly effects on production of milk are displayed by aggregating data values on a cow-relative timeline. Basic statistics on data values are dynamically calculated and a density plot is used to quantify the reliability of a dataset. A qualitative expert user study conducted with animal researchers shows that the system is an important means to identify anomalies in data collected and to understand dominant data patterns, such as clusters of samples and outliers. The evaluation is complemented by a case study with two datasets from the field of dairy science.

Paper Nr: 9
Title:

Analyzing Intrinsic Motion Textures Created from Naturalistic Video Captures

Authors:

Angus Graeme Forbes, Christopher Jette and Andrew Predoehl

Abstract: This paper presents an initial exploration of the plausibility of incorporating subtle motions as a useful modality for encoding (or augmenting the encoding of) data for information visualization tasks. Psychophysics research indicates that the human visual system is highly responsive to identifying and differentiating even the subtlest motions intrinsic to an object. We examine aspects of this intrinsic motion, whereby an object stays in one place while a texture applied to that object changes in subtle but perceptible ways. We hypothesize that the use of subtle intrinsic motions (as opposed to more obvious extrinsic motion) will avoid the clutter and visual fatigue that often discourages visualization designers from incorporating motion. Using transformed video captures of naturalistic motions gathered from the world, we conduct a preliminary user study that attempts ascertains the minimum amount of motion that is easily perceptible to a viewer. We introduce metrics which allow us to categorize these motions in terms of flicker (local amplitude and frequency), flutter (global amplitude and frequency), and average maximum contrast between a pixel and its immediate neighbors. Using these metrics (and a few others), we identify plausible ranges of motion that might be appropriate for visualization tasks, either on their own or in conjunction with other modalities (such as color or shape), without increasing visual fatigue. Based on an analysis of these initial preliminary results, we propose that the use of what we term “intrinsic motion textures” may be a promising modality appropriate for a range of visualization tasks.

Paper Nr: 28
Title:

Visualising Java Coupling and Fault Proneness

Authors:

P. Rosner, M. Child and S. Counsell

Abstract: In this paper, a tool is described for visualising the Coupling Between Objects (CBO) metric for Java systems, decomposing it into coupling collaborators and using colour to denote the object-oriented mechanisms at work for each couple. The resulting visualisation is also envisaged to be useful for general program comprehension and is integrated into Java development in the Eclipse IDE. Evidence is also given that the visualisation may help detect classes tending to be less fault-prone than would be expected from inspection of their CBO values alone.

Paper Nr: 32
Title:

Reconstructing Conimbriga - Digital Cantaber

Authors:

César Ferreira, Nuno Rodrigues, Alexandrino Gonçalves and Virgílio Hipólito-Correia

Abstract: Being the ancient cultural heritage structures and artefacts so full of detail and relevance to studies related to our past, it would be desirable to have precise virtual replicas that could be freely explored without endangering important pieces of history. However, the costs to produce three-dimensional environments are sometimes discouraging, due to the significant cost of some 3D authoring tools, and to the time necessary to manually produce the models. This paper presents a low-cost alternative to the classic manual modelling process, towards the production of highly detailed virtual models, by using open source software and a low-cost moving depth camera. The ambition is the dissemination of our cultural heritage legacy, making it accessible, not only to experts, but also to the general public, without requiring any high performance hardware, authoring software or professional 3D skills. For the visualization, our virtual reconstructions will be available through a three-dimensional live model viewer, based on recent technologies such as HTML5 and WebGL. Those may be triggered from a wide variety of device, contributing in this way to a true democratization of history knowledge. The proposed approach was applied to create a virtual model of the so-called “House of Cantaber”.

Paper Nr: 38
Title:

A Geological Metaphor for Geospatial-temporal Data Analysis

Authors:

Tom Liebmann, Patrick Oesterling, Stefan Jänicke and Gerik Scheuermann

Abstract: To provide visual access to geospatial-temporal data, existing systems usually highlight the data’s spatial, temporal and topical distribution individually in separated, but linked views. Because this design often complicates queries that concern multiple data aspects and also involves more user interaction, in this paper, we present a geological metaphor that aims to combine relations between orthogonal data aspects. We describe how our adopted landscape metaphor intuitively depicts global and local relationships based on its surface, glyph augmentation and inner sediment structure. We validate the geological metaphor with case studies, compare it with existing systems and describe how it can be integrated into those as an alternative map view.

Paper Nr: 41
Title:

Force Directed Flow Map Layout

Authors:

Alberto Debiasi, Bruno Simões and Raffaele De Amicis

Abstract: A flow map is a thematic map that is been used to emphasize the spatial pattern of one or more geographic attributes. Although this kind of thematic maps is often drawn by hand, a few automatic computer algorithms exist. In this research paper, we proposed a novel algorithm for the automatic generation of flow maps that is theoretically grounded on physics’ laws to describe the motion and force of attraction or repulsion between points. Properties associated to these laws are then used to merge different flows, as well as for the improvement of the maps’ visual quality. Finally, we evaluate our work by generating a set of flow maps and by doing a comparison with flow maps produced by existing algorithms.

Paper Nr: 43
Title:

Visualization and Clustering of Online Book Reviews

Authors:

Shiaofen Fang, Lanfang Miao and Eric Lin

Abstract: Online user reviews of products, movies, books, etc. have been an important source of information for applications such as social networking, online retail, and sentiment analysis. In this paper, we present a novel visualization tool for analysing and visualizing online book reviews. Using text mining techniques, nontrivial features (tags) are identified on the text data extracted from the online reviews. These keyword tags are used to cluster both the books and the readers based on global tag similarities. Two different visualization methods are proposed: parallel coordinate views and 3D correlative cluster views. The parallel coordinate visualization provides a flat view of the tag distributions to reveal clustering patterns. A novel 3D corrective visualization technique is developed to visually represent the correlations of reader clusters and book clusters. These visualization techniques can also be applied to other types of online text data in social networks and web commerce.

Paper Nr: 51
Title:

A Sketch of a Theory of Visualization

Authors:

Randy Goebel

Abstract: A picture results from a possibly multi-layer transformation of data to a visual vocabulary in which humans can draw inferences about the original data. The goal of this visualization process is to expose relationships amongst the data that are otherwise difficult to find, or only emerge by the process of the transformation. In case of the former kind of inference (confirming a relationship that did exist but was not obvious), visualization provides a kind of inferential amplifying effect. In the case of the latter (exposing new data relationships), visualization provides an inductive mechanism to create hypotheses not manifest in the original data. In this regard, the creation of pictures from data is about data compression, which is naturally a kind of machine learning. Just as statistical concepts like average and standard deviation provide a measure on properties of a set of numbers, so too does visualization provide a kind of ``measure'' on data compressed to a visual vocabulary presented as a picture. Our position is that visualization is about the (potentially multi-step, multi-layered) transformation of data to pictures, and that ever such transformation must make choices about what kinds of relations {\it to preserve}, and what kinds of data artifacts {\it to avoid} in each such transformation. Like a chain of formal inference, conclusions following from the end result (the picture) are determined by what each transformation in the inference chain is intended to accomplish. We argue that the visualization of large data sets, too large to inspect directly, requires a rigorous theory of how to transform data to pictures, so that the scientists as observers can be assured that inferences drawn from the pictures are either confirmable in the detailed data, or at least plausible hypotheses which can be further pursued by seeking further data (evidence).

Area 2 - General Data Visualization

Full Papers
Paper Nr: 15
Title:

Eye-tracking Investigation During Visual Analysis of Projected Multidimensional Data with 2D Scatterplots

Authors:

Ronak Etemadpour, Bettina Olk and Lars Linsen

Abstract: A common strategy for visual encoding of multidimensional data for visual analyses is to use dimensionality reduction. Each multidimensional data point is projected to a 2D point using a certain strategy for the 2D layout. Many layout strategies have been proposed addressing different objectives and targeted at distinct domains and applications. The resulting projected information is typically displayed in form of 2D scatterplots. The user’s perspective such as the role of visual attention and guidance of attention for a respective layout and task has not been addressed much. It is the goal of this work to investigate, how characteristics in the layout affect the cognitive process during task completion. Eye trackers are an effective means to capture visual attention over time. We use eye tracking in a user study, where we ask users to perform typical analysis tasks for projected multidimensional data such as relation seeking, behavior comparison, and pattern identification. Those tasks often involve detecting and correlating clusters. To understand the role of point density within clusters, cluster sizes, and cluster shapes, we first conducted a study with synthetic 2D scatterplots, where we can set the respective properties manually. We evaluate how changing various parameters affect the visual attention pattern and correlate it to the correctness of the answer. In a second step, we conducted a study where the users were asked to complete tasks on real-world data with different characteristics (image collection and document collection) that are visualized using a selection of different dimensionality reduction algorithms. We transfer the insight obtained from synthetic data to investigate the decision making with real-world data. Gestalt laws can be applied to the layout structure. We examine how certain layout techniques produce certain characteristics that change the visual attention pattern. We draw some conclusions on how different projection methods support or hinder decision making leading to respective guidelines.

Paper Nr: 23
Title:

VizClick - Visualizing Clickstream Data

Authors:

Rajat Kateja, Amerineni Rohith, Piyush Kumar and Ritwik Sinha

Abstract: Clickstream data is ubiquitous in today’s web-connected world. Such data displays the salient features of big data, that is, volume, velocity and variety. As with any big data, visualizations can play a central role in making sense and generating hypotheses from such data. In this paper, we present a systematic approach of visualizing clickstream data. There are three basic questions we aim to address. First, we explore the interdependence between the large number of dimensions that are measured in clickstream data. Next, we analyze spatial aspects of data collected in web-analytics. Finally, the web designers might be interested in getting a deeper understanding of the website’s topography and how browsers are interacting with it. Our approach is designed for business analysts, web designers and marketers; and helps them draw actionable insights in the management and refinement of large websites.

Short Papers
Paper Nr: 14
Title:

Suggesting Visualisations for Published Data

Authors:

Belgin Mutlu, Patrick Hoefler, Gerwald Tschinkel, Eduardo Veas, Vedran Sabol, Florian Stegmaier and Michael Granitzer

Abstract: Research papers are published in various digital libraries, which deploy their own meta-models and technologies to manage, query, and analyze scientific facts therein. Commonly they only consider the meta-data provided with each article, but not the contents. Hence, reaching into the contents of publications is inherently a tedious task. On top of that, scientific data within publications are hardcoded in a fixed format (e.g. tables). So, even if one manages to get a glimpse of the data published in digital libraries, it is close to impossible to carry out any analysis on them other than what was intended by the authors. More effective querying and analysis methods are required to better understand scientific facts. In this paper, we present the web-based CODE Visualisation Wizard, which provides visual analysis of scientific facts with emphasis on automating the visualisation process, and present an experiment of its application. We also present the entire analytical process and the corresponding tool chain, including components for extraction of scientific data from publications, an easy to use user interface for querying RDF knowledge bases, and a tool for semantic annotation of scientific data sets.

Paper Nr: 18
Title:

Role of Human Perception in Cluster-based Visual Analysis of Multidimensional Data Projections

Authors:

Ronak Etemadpour, Robson Carlos da Motta, Jose Gustavo de Souza Paiva, Rosane Minghim, Maria Cristina Ferreira de Oliveira and Lars Linsen

Abstract: Visualization of high-dimensional data requires a mapping to a visual space. Whenever the goal is to preserve similarity relations, multidimensional projections or other dimension reduction techniques are commonly used to project high-dimensional data point to a 2D point using a certain strategy for the 2D layout.Typical analysis tasks for projected multidimensional data do not necessarily match the expectations of human perception. Learning more about the effectiveness of projection layouts from a users perspective is an important step towards consolidating their role in supporting visual analytics tasks. Those tasks often involve detecting and correlating clusters. To understand the role of orientation and cluster properties of size, shape and density, we first conducted a study with synthetic 2D scatter plots, where we can set the respective properties manually. Then we picked five projection methods representative of different approaches to generate layouts of high dimensional data for two domains, image and document data. The users were asked to identify the clusters on real-world data and answers to questions were compared for correctness against ground truth computed directly from the data. Our results offer interesting insight on the use of projection layouts in data visualization tasks.

Posters
Paper Nr: 12
Title:

IPFViewer - A Visual Analysis System for Hierarchical Ensemble Data

Authors:

Matthias Thurau, Christoph Buck and Wolfram Luther

Abstract: Analyzing ensemble data is very challenging due to the complexity of the task. In this paper, we describe IPFViewer, a visual analysis system for ensemble data, that is hierarchical, multidimensional and multimodal. The exemplary data set comes from a steel production facility and comprises data about their melting charges, samples and defects. Our system differs from existing ones in that it encourages the usage of side-by-side visualization of ensemble members. Besides trend analysis, outlier detection and visual exploration, side-by-side visualization of detailed ensemble members enables rapid checking for repeatability of single ensemble member analysis results. IPFViewer supports the following data interaction methods: Hierarchical sorting and filtering, reference data selection, automatic percentile selection and ensemble member aggregation, while the focus for visualization is on small multiples of multiple views.

Paper Nr: 24
Title:

Visual Analysis of Perceptual and Cognitive Processes

Authors:

Michael Raschke, Tanja Blascheck, Marianne Richter, Tanja Agapkin and Thomas Ertl

Abstract: The success of visualization techniques depends on their support of perceptual and cognitive processes to perceive the graphically represented information. Apart from measuring accuracy rates of correctly given answers and completion times in user studies, eye tracking experiments provide an additional technique to analyze perceptual and cognitive processes of visual tasks. This paper presents an interdisciplinary approach for studying structures of scan paths by visual means. We propose to annotate graphical elements with semantic information. This annotation allows us to analyze the fixation sequences on these annotated graphical elements with respect to reading processes, visual search strategies, and visual reasoning.

Paper Nr: 36
Title:

GCLViz: Garbage Collection vs. Latency Visualization

Authors:

Chihua Ma, Stanislav Liberman and Haifeng Zheng

Abstract: This paper proposes a method that creates a multi-view interactive visualization that allows users to explore connections between garbage collection (GC) generated by Java Virtual Machine (JVM) and latency in applications used in financial transactions. With this tool users can explore large collections of GC and latency events, easily identify important events, and subsequently focus on the relationships and details of such events without losing the “big picture” perspective on the events as a whole. We discuss the impact of this tool on controlling the effects of GC on latency and variability in financial trades with an exchange.

Area 3 - Spatial Data Visualization

Full Papers
Paper Nr: 2
Title:

Fine-Grained Provenance of Users’ Interpretations in a Collaborative Visualization Architecture

Authors:

Aqeel Al-Naser, Masroor Rasheed, Duncan Irving and John Brooke

Abstract: In this paper, we address the interpretation of seismic imaging datasets from the oil and gas industry—a process that requires expert knowledge to identify features of interest. This is a subjective process as it is based on human expertise and thus it often results in multiple views and interpretations of a feature in a collaborative environment. Managing multi-user and multi-version interpretations, combined with version tracking, is challenging; this is supported by a recent survey that we present in this paper. We address this challenge via a data-centric visualization architecture, which combines the storage of the raw data with the storage of the interpretations produced by the visualization of features by multiple user sessions. Our architecture features a fine-grained data-oriented provenance, which is not available in current methods for visual analysis of seismic data. We present case studies that present the use of our system by geoscientists to illustrate its ability to reproduce users’ inputs and amendments to the interpretations of others and the ability to retrace the history of changes to a visual feature.

Short Papers
Paper Nr: 21
Title:

Visualizing Large Scale Vehicle Traffic Network Data - A Survey of the State-of-the-art

Authors:

H. W. A. S. Gondim, H. A. D. do Nascimento and D. Reilly

Abstract: Analyzing and improving large urban traffic networks is a difficult process due to complex interrelationships between the many variables that impact vehicle traffic behavior. Information visualization techniques can facilitate the tasks of analyzing large amounts of data and of exploring potential solutions to practical traffic problems. Surprisingly, there is a relative lack of investigation focused on how information visualization techniques should be applied and adapted to the field of Traffic Engineering. This paper presents an overview of what has been done on this topic by reviewing the use of information visualization in traffic systems over the years, and highlighting the current state-of-the-art by focusing on several innovative pieces of research. We provide a classification of the reviewed work and identify areas that have been understudied.

Posters
Paper Nr: 10
Title:

Uncertainty Estimation and Visualization of Wind in Weather Forecasts

Authors:

Bård Fjukstad, John Markus Bjørndalen and Otto Anshus

Abstract: The Collaborative Symbiotic Weather Forecasting system, CSWF, let individual users do on-demand small region, short-term, and very high-resolution forecasts. When the regions have some overlap, a symbiotic forecast can be produced based on the individual forecasts from each region. Small differences in where the center of the region is located when there is complex terrain in the region, leads to significant differences in the forecasted values of wind speed and direction. These differences reflect the uncertainty of the numerical model. This paper describes two different ways of presenting these differences using a traditional map based approach on a laptop and a display wall, and an augmented reality approach on a tablet. The approaches have their distinct advantages and disadvantages depending on the actual use and requirements of the user.

Paper Nr: 16
Title:

Visual Analytics of Multi-sensor Weather Information - Georeferenciation of Doppler Weather Radar and Weather Stations

Authors:

Aitor Moreno, Andoni Galdós, Andoni Mujika and Álvaro Segura

Abstract: This work presents a geovisual tool which integrates and georeferences data coming from some of the weather instruments installed in the Basque Country: a Doppler weather radar and the weather station network composed of around 100 multi-sensors stations (temperature, precipitation, wind...). The visualization of the raw data coming from the weather radar is based on the generation of a set of 3D textured concentric cones (one per elevation scan). The resulting 3D model is then integrated in the 3D digital terrain of the Basque Country. For the weather stations, we have provided a Kriging based interpolation method to produce textures from the scalar data measured at the weather stations. These textures are then mapped in the same 3D digital terrain as before. The integrated visualization of the weather information enhances the understanding of the data. To illustrate the proposed methods a use case is provided: matching the precipitation measured at ground level with the radar scans.

Paper Nr: 26
Title:

Interactive Stream Surface Placement - A Hybrid Clustering Approach Supported by Tree Maps

Authors:

M. Edmunds, R. S. Laramee, R. Malki, I. Masters, Y. Wang, G. Chen, E. Zhang and N. Max

Abstract: The ability of a CFD engineer to study, capture, and visualise 3D flow simulation data is a challenge. Stream surfaces are a useful tool for visualising 3D flow because of their ability to convey many field attributes from their structure. It is important that the CFD engineer can interact with, and examine specific characteristics of the CFD data. We introduce an interactive, cluster based stream surface placement strategy for structured and unstructured CFD data. A two-phase hybrid clustering algorithm is used to visualise interesting subsets of the flow. An interactive tree map interface provides a visual overview and enables interactive selection of cluster details corresponding to interesting features of the data at which to place stream surfaces. We demonstrate the performance and effectiveness of our interactive framework on a range of flow simulations and provide domain expert evaluation of the results, providing valuable insight for the CFD engineers.

Paper Nr: 40
Title:

Hardware-Accelerated Attribute Mapping for Interactive Visualization of Complex 3D Trajectories

Authors:

Stefan Buschmann, Matthias Trapp, Patrick Lühne and Jürgen Döllner

Abstract: The visualization of 3D trajectories of moving objects and related attributes in 3D virtual environments represents a fundamental functionality in various visualization domains. Interactive rendering and visual analytics of such attributed trajectories involves both conceptual questions as well as technical challenges. Specifically, the mapping of trajectory attributes to rendering primitives and appearance represents a challenging task in the case of large data sets of high geometric complexity. There are various visualization approaches and rendering techniques considering specific aspects of these mappings to facilitate visualization and analysis of this kind of data. To solve the underlying general mapping problem efficiently, we developed an approach that uses and combines diverse types of visualizations, rather than being tailored to a specific use case. This paper describes an interactive rendering system for the visualization of 3D trajectories that enables the combinations of different mappings as well as their dynamic configuration at runtime. A fully hardware-accelerated implementation enables the processing of large sets of attributed 3D trajectories in real-time.