Accepted Workshops

VAST Challenge
Location: SCENE C


The 2014 VAST Challenge featured a set of interrelated challenges that required participants to analyze multiple diverse and conflicting datasets to solve a mystery. Participants developed solutions using realistic but artificial geospatial and geotemporal data, transaction data, email header data, unstructured text, and streaming microblog data. A record 73 entries were received from 13 countries.
The VAST Challenge workshop brings together organizers, participants, and conference attendees to discuss the innovative submissions to this year’s challenge. The workshop will feature sessions dedicated to each of the three mini-challenges, as well as the grand challenge. The 2014 award winners and honorable mention winners will present their submissions. In addition, the meeting will feature a poster session and a participant feedback session. This workshop is open to all IEEE VIS attendees.


Visualizing Electronic Health Record Data
08:30-12:10 ---- Location: SCENE B
14:00-18:00 ---- Location: (Off site) Telecom ParisTech

Catherine Plaisant, Silvia Miksch, Theresia Gschwandtner, Sana Malik


Electronic Health Record (EHR) databases contain millions of patient records including events such as diagnoses, test results, or medication prescriptions. These records are an invaluable data source for clinical research and improvement of clinical quality, as they provide longitudinal health information about patient populations. The use of EHR databases could be dramatically improved if easy-to-use interfaces allowed clinical researchers and quality improvement analysts to explore complex patterns in order to build and test hypotheses regarding the benefits, risks, and appropriateness of treatments or medication regimens.
Novel strategies in information visualization and visual analytics are needed. The interest in this topic is growing at very rapid pace and is very interdisciplinary by nature, both in term of field (medicine and computer science) but also research environment (academic research as well as industry and government agencies). Because of the European location of the conference, we have a unique opportunity to create bridges and explore new collaborations between groups that would have never met otherwise.


DECISIVe: Dealing with Cognitive Biases in Visualisations
08:30-12:10 ---- Location: AUDITORIUM
14:00-18:00 ---- Location: (Off site) Telecom ParisTech

Geoffrey Ellis, David Peebles, Donald Kretz, Gaëlle Lortal


Our inherent reliance on mental shortcuts, or heuristics, sometimes results in deviations in judgment from what rational decision models would predict. These deviations are known as cognitive biases. Heuristics allow us to make “good enough” decisions without expending all of our cognitive effort on the task, however, in critical decision environments, “good enough” is often NOT good enough. Visualisation tools are increasingly adept at making sense of complex data, but researchers who study cognitive biases have come to realise that the quality of decisions made with these tools are often impaired because tool designers fail to address how heuristics and biases operate in a human-computer interactive setting. The aim of this workshop is to bring together a wide range of researchers and developers from domains such as information visualisation, visual analytics and cognitive psychology to explore some of the ways in which biases impact user performance and share ideas and experiences about practical ways to reduce or overcome these potentially harmful effects in the systems we build. 


Towards An Open Visualization Literacy Testing Platform
80:30-12:10 ---- Location: FORUM A,B,C
14:00-18:00 ---- Location: (Off site) Telecom ParisTech

Sung-Hee Kim, Jeremy Boy, Sukwon Lee, Ji Soo Yi, and Niklas Elmqvist


We propose a hands-on workshop where participants will learn about and discuss visualization literacy by actually designing and evaluating questions for a visualization literacy measure. Though the value of information visualization is becoming apparent to a broad audience, visualization researchers often acknowledge that people have different levels of understanding of visualization techniques. In other words, our understanding of how users interpret visualizations has not caught up with design and technical developments, and even the concept of visualization literacy is still debated. Different domains of research, such as mathematics education, cognitive science, and psychology, have been approaching this problem within their domain. We believe that researchers in information visualization and visual analytics should lead the effort in defining the concept, and in creating valid and practical measurement tools. The goal of our workshop is to take a step in this direction by developing a better understanding of visualization literacy, identifying possible metrics for evaluation, and raising new questions for future research through the design and evaluation of visualization literacy tests. The outcome of our workshop will be a participatory web-platform for collectively created visualization literacy tests and questionnaires that can directly be used by researchers in our community. 


3DVis@IEEEVIS2014: Does 3D really make sense for Data Visualization?
08:30-12:10 ---- Location: SCENE D,E,F
14:00-18:00 ---- Location: (Off site) Telecom ParisTech

Thierry Duval, Pascale Kuntz, Jérôme Royan, Wolfgang Stuerzlinger, Gilles Venturini


The aim of this workshop is to discuss 3D display and 3D user interfaces for scientific and information visualization and Visual Analytics. As such, it starts by analyzing the state of the art in 3D visualization and associated interaction. We also review how 3D displays and user interfaces match human perceptual, cognitive and motor capabilities. We then go to the heart of the 2D vs. 3D debate. On the one hand, 3D technologies provide additional options for display and interaction, which broadens the space of potential solutions considerably. Consequently, many 3D visualization approaches have been implemented. The recent introduction of low-cost virtual and augmented reality hardware further paves the way towards a deeper exploration of the design space for 3D visualization. Moreover, better consistency in terms of dimensions between the visualization and interaction may also improve the immersion and navigation in huge amounts of data. On the other hand, the experimental results from studies that investigate the differences between 2D and 3D visualization interfaces provide no real evidence for the claimed benefits of 3D technologies. Thus, the current body of literature provides at best inconclusive guidance on the benefits of 3D approaches. Moreover, a deep consideration of human abilities and capabilities identifies a potential mismatch with full 3D user interfaces. Yet, there may be a bias in the evaluations, as the prevalent visualization and interaction techniques in all areas of visualization are based on non-immersive technologies. In the workshop, we will present evidence for both sides of this debate and finish up with directions for future work.


Death of the Desktop – Envisioning Visualization without Desktop Computing
Location: SCENE B

Yvonne Jansen, Petra Isenberg, Jason Dykes, Sheelagh Carpendale, Sriram Subramanian, Daniel Keefe


The Desktop computer is dead. Monitors sit on desks, unplugged - hosting layers of Post-It notes or gathering dust as a retro emergency low-light mirror. Visualization is colourful, big, tangible, nosy, interactive, compelling and everywhere. It supports all sorts of creative activity and is key to problem solving in education, science, government and industry.
But how?
What is your 'imagined future' for visualization?
We will be exploring possible visualization scenarios with short but rich scenarios in which designers, practitioners and researchers creatively explore opportunities for 'beyond-the-desktop' visualization. We will be discussing these and using them to develop the community’s perspective on the future of VIS.


Visualization for Predictive Analytics

E. Bertini, A. Perer, R. Maciejewski, J. Sun


One of the surprising facts of much current visualization research is that prediction does not often play a significant role. Most visualization research seems to focus exclusively on data analysis and presentation, with little support for predictive analytics and the numerous models researchers have developed for this purpose. Upon reflection, this comes as a surprise as many scientific endeavors and many business problems are mostly concerned with prediction. Looking more closely at the recent advancements (and tremendous popularity) of Data Science, one may recognize that the vast majority of problems addressed involve some form of prediction and modelling. Notable examples are: prediction of drug effectiveness in drug development, prediction of diseases in healthcare, prediction of crime in city management.
Our goals are to increase the awareness about this interesting opportunity for visualization research, collect and compare examples of existing and ongoing research in this area and to for visual analytics researchers. The workshop will allow participants to showcase their existing research and ideas and to learn and reflect on the latest advances in visualization of predictive models.  


Provenance for Sensemaking
Location: SCENE B

Kai Xu, Simon Attfield, T.J. Jankun-Kelly


During complex sensemaking and analysis task, it can be valuable to maintain a history of the processes and transformations involved - referred to as ‘provenance’ information. Provenance information can be a resource for "reflection-in-action" during analyses, for supporting planning and reframing of objectives and scope. It can also be a resource after the event, supporting the interpretation of claims, audit, accountability or training.
There has been considerable work on capturing and visualizing of ‘data provenance’, which focuses on data collection and computation, and ‘analytic provenance’, which captures the interactive data exploration process. However, there is limited work of utilizing these provenance information to support sensemaking, in terms of improving its efficacy and avoid pitfalls such as data quality issue and human bias.
This workshop aims to bring together researchers involved in visual analytics and various aspects of sensemaking to consider emerging positions, questions, and findings related to the capture, processing, representation and use of provenance information to support complex sensemaking tasks. The emphasis is on discussion and collaboration, with a goal to produce a paper describing the state-of-the-art of provenance for sensemaking after the workshop.


Location: SCENE B

Rahul C. Basole, Steven Drucker, Jörn Kohlhammer, Jack van Wijk


Enterprises across all industries are inundated by an accelerating tsunami of business data. Converting these data and insights into better business outcomes is a pressing and strategic challenge for many managers, decision makers, and executives. While there is a growing interest in business analytics, advanced information visualization and visual analytic tools have lagged behind. This is in part explained by the plethora and complexity of design and implementation challenges presented by diverse business users, data, and contexts. This workshop aims to bring together researchers and practitioners interested in in presenting, exploring, and understanding complex business issues through visualization and visual analytics.