Welcome! I am Fabian Klute, currently I am a PostDoctoral researcher with the Geometric Computing group at Utrecht University. I come from Landau/Palatina in Germany and studied my Bachelor's and Master's degree in Karlsruhe. Recently, I finished my PhD at the TU Wien with Martin Nöllenburg.My main research interests are in the area of graph drawing and automated cartography. At the moment I am very interested in map labeling, content generation, and confluent drawings, as well as parametrized complexity and its application to various geometric topics.
Short CV:
April 2020 – Present
PostDoc in the Geometric Computing group at Utrecht University with Prof. Marc J. van Kreveld
April 2016 – March 2020
PhD Student in the Algorithms and Complexity Group at TU Wien under the supervision of Prof. Martin Nöllenburg
May 2015 – December 2015
Gap months, some impressions can be found here
April 2013 – April 2015
Master of Science at the Karlsruher Institute of Technology (KIT)
October 2009 – March 2013
Bachelor of Science at the Karlsruher Institute of Technology (KIT)
I enjoy a lot to participate in the yearly contest held at the Graph Drawing conference. Over the years I created several posters.
GD'18 – 4th place
Submission for the GD Contest 2018 together with Carlos HidalgoToscano and Irene Parada. The presented data is a graph based on Game of Thrones.
GD'17 – 1st place
Submission for the GD Contest 2017 together with Irene Parada. The data is taken from the human metabolism project. In order to filter the information, we thin out the graph in a preprocessing step. For every reaction we keep only the lowestdegree reactant and product. Isolated and degree one nodes are removed and only the biggest component is kept. Additionally we identify nodes corresponding to different biological compartments (e.g. co2[c], co2[e] are contracted into one co2 node).
To generate the layout we add edges to keep the subsystems closer together. The size of the nodes depends on its betweenes in the reduced graph. Then the layout is calculated using the FruchtermanReingold algorithm. Finally we color the nodes based on subsystems. We distinguish two big groups of subsystems, the communication or transportation ones and the proper ones. As the name suggests the first kind mainly connects the proper subsystems. Based on this we categorize nodes into five cases, depending on if they are inside a proper subsystem, on its boundary, in a transport system, on the boundary between two different transport systems, or between two proper subsystems.
To generate the layout we add edges to keep the subsystems closer together. The size of the nodes depends on its betweenes in the reduced graph. Then the layout is calculated using the FruchtermanReingold algorithm. Finally we color the nodes based on subsystems. We distinguish two big groups of subsystems, the communication or transportation ones and the proper ones. As the name suggests the first kind mainly connects the proper subsystems. Based on this we categorize nodes into five cases, depending on if they are inside a proper subsystem, on its boundary, in a transport system, on the boundary between two different transport systems, or between two proper subsystems.
GD'16 – 1st place
Submission for the GD Contest 2016. The data is an experpt of the data surfaced in the panama papers. Graph and layout were produced by exploring the graph with yed, R, whatever C program I had lying around. I focused mainly on finding sets of strongly connected vertices. I noticed a group of eleven vertices which are connected by an undirected circle and on top are almost all connected to each other in some way. I extracted this group and made one visualization for them in Ipe (left half). Among the leftover vertices a lot of them have no incoming edges so I plotted them in treemaps using R. These treemaps I then used as vertices and connected them with the other remaining vertices with incoming edges that were not part of the above mentioned group (right half). All layouting was done in Ipe by hand with the help of a couple of python scripts generating e.g. the legend.
Here you can find the teaching I did during my time at TU Wien.
SS19
WS18/19
SS18
WS17/18
SS17
WS16/17
SS16
Algorithmen und Datenstrukturen. Exercise lessons. Redesign of the programming part of the course. [Link]
Graph Drawing Algorithms. Exercise lessons and coordination. [Link]
Algorithmic Geometry. Exercise lessons and designing the exercise sheets. [Link]
Seminar in Algorithms Graphs and Geometry. [Link]
Algorithmen und Datenstrukturen. Exercise lessons and programming exercises. [Link]
Graph Drawing Algorithms. Exercise lessons and coordination. [Link]
Algorithmic Geometry. Exercise lessons and designing the exercise sheets. [Link]
Heuristic Optimization Techniques. Help with exercises. [Link]
Seminar in Algorithms Graphs and Geometry. [Link]
Algorithmic Geometry. Exercise lessons and designing the exercise sheets. [Link]
Seminar in Algorithms. [Link]
Algorithmen und Datenstrukturen. Exercise lessons. [Link]

Extending to 1Plane Drawings.Abstracts of the XVIII Spanish Meeting on Computational Geometry (EGC'19), 30–33, 2019.

Maximizing Ink in Partial Edge Drawings of kPlane Graphs.27th International Symposium on Graph Drawing and Network Visualization (GD'19), volume 11904 of LNCS, 323–336, 2019.

Minimizing Crossings in Constrained TwoSided Circular Graph Layouts.Journal of Computational Geometry, 10(2), 2019.Minimizing Crossings in Constrained TwoSided Circular Graph Layouts.34th International Symposium on Computational Geometry (SoCG'18), volume 99 of LIPIcs, 53:1–53:14, 2018.

Serealisierung Und Deserialisierung von Mustern in Zellularautomaten.Bachelor's Thesis, Karlsruhe Institute of Technology (KIT), 2012.