I am a PhD candidate at the Eindhoven University of Technology within the Dept. Math. and Computer Science, Visualization group of prof.dr.ir. Jarke J. van Wijk. My position is supported by SynerScope BV, a spin-off company of the visualization group providing tools and techniques for fraud and forensics building further upon the work of Danny Holten (one of my supervisiors). In the context of my PhD I perform research on visual analytics for dynamic networks. I obtained my MSc. / ir. degree (with honors) in Computer Science and Engineering also from the Eindhoven University of Technology in 2011.
In my spare time I like to code on hobby projects, play Magic the Gathering, toy around with microcontrollers trying to build robots, watch movies, go skating if the weather allows for this or listen to something I consider music.
The main research question for my PhD project is:
‘‘how to enable people to obtain insight in very large amounts of transaction data using a combination of automated and visual methods?’’
Current methods are based on purely visual analysis. The main challenge here is scalability; these visual methods fall short, for instance because of limited screen resolution. Also, current state of the art as well as industry standards are mainly based on purely automated methods such as rule based retrieval. However, these automated methods fall short due to aggregated results and loss of context. Furthermore, no exploration of the data is possible for instance to find new patterns not earlier defined by a rule. We believe a right combination of both visual and automated methods supports users in the analytical reasoning process to detect anomalies and obtain new insights from the data. The application domains are financial transactions, accounting data, network data or, in general, any domain data that can be modeled as a dynamic network. The aims of the project are to develop algorithms and interactive techniques for dynamic network analysis and visualization. The main goals are to study, develop and evaluate in close cooperation with industry and domain experts:
- selection of suitable automated methods,
- development of interaction methods that enable users to specify what they find interesting in a natural and intuitive manner,
- development of new visual metaphors to show the data, in raw as well as compressed form (after analysis),
- development of methods to enable users to store and present the results of their analysis.
Instructor2IV35 2013: Visualization
2IV60 2013: Computer Graphics
2IV10 2012: Computer Graphics
Student assistant2IO60 2009: Design Based Learning 3.1: Development of a 3D network game
2IO60 2008: Design Based Learning 3.1: Development of a 3D network game
Exploration and Analysis of Massive Mobile Phone Data: A Layered Visual Analytics approach, Stef van den Elzen, Jorik Blaas, Danny Holten, Jan-Kees Buenen, Jarke J. van Wijk, Robert Spousta, Anna Miao, Simone Sala, Steve Chan in Proceedings of the 3rd International Conference on the Analysis of Mobile Phone Datasets, (NetMob 2013), Cambridge, MA, May 2013. (best visualization award) to appear
Visualizing group dynamics signals, Stef van den Elzen, Henri Hambartsumyan, Erik Kuijpers in Measuring Behavior '10: Proceedings of the 7th International Conference on Methods and Techniques in Behavioral research