Bio

Felix is a research assistant in the ExCELL project. He has a Bachelor degree in Geography and a Master degree in Geoinformation Science. After his studies he has worked as a database developer for managing 3D city models. Aside from his work at the Beuth University he gives lectures at universities on geodatabases.

 

Research Interests

  • Spatial data management and storage (RDBMS, NoSQL, HDFS)
  • Spatio-temporal statictics
  • Linear Referencing
  • Machine learning in GIS
  • 3D city modelling
  • Teaching students the greatness of PostGIS

 

Current Research

  • Short-term traffic forecasting including spatial information using different methods (STARIMA, SVM, Artificial Neural Networks)
  • How to handle missing data in time series
  • Scalable geostatistical algotrithms running on top of Hadoop
  • Open data model for traffic information in OpenStreetMap
  • Mobile databases using the OGC Standard GeoPackage

Publifications

  • Kunde, F.; Sauer, P. (2017). pgMemento–A Generic Transaction-Based Audit Trail for Spatial Databases. 15th International Symposium on Spatial and Temporal Databases 2017, 416-420. Springer.
  • Sauer, P.; Pieper, S.; Kunde, F.; Waterloo, S. (2017). Correction of Incomplete and Incorrect Time Series for Traffic Prediction. In: Milosz, M.; Tukeyev, U. (Eds.). Varia Informatica 2017. 217-226. PIPS.
  • Pieper, S.; Kunde, F.; Waterloo, S.; Sauer, P. (2017). Bereinigung unvollständiger und fehlender Messreihen für Verkehrsvorhersagen. In: Research Day 2017. Beuth-Hochschule Berlin.
  • Kunde, F.; Hartenstein, A.; Sauer, P. (2017). Spatio-Temporal Traffic Flow Forecasting on a City-Wide Sensor Network. In: Ivan, I.; Horák, J; Inspektor, T. (Eds.). Dynamics in GISciene. Proceedings of GIS Ostrava, 253-265. Springer.
  • Kunde, F; Hartenstein, A.; Pieper, S.; Sauer, P. (2017). Traffic prediction using a Deep Learning paradigm. 1st International Workshop on Big Geo Data Quality and Privacy (BIGQP).
  • Agoub, A.; Kunde, F.; Kada, M. (2016). Potential of Graph Databases in Representing and Enriching Standardized Geodata. In: Dreiländertagung der DGPF, der OVG und der SGPF - Proceedings. Bern. Volume: 25.
  • Kunde, F.; Pieper, S.; Sauer, P. (2016): ExCELL - Eine Plattform für die Vorhersage und Analyse von Verkehrsdaten. In: Research Day 2016. Beuth-Hochschule Berlin.
  • Kunde, F.; Nagel, C.; Herrerruela, J.; Ross, L; Kolbe, T.H. (2013). 3D-Stadtmodelle in PostGIS mit der 3D City Database. In: Tagungsband der FOSSGIS-Konferenz 2013. Rapperswil.

 

Presentations

 

Supervised Student Theses and Projects

  • Hozan Ibrahim: Anomaly detection in traffic data
  • Anja Langer: Database design with Geopackage
  • Anass Benabdennbi: Performance comparison between PostGIS, Exasol and Impala
  • Minh Hoang Dang: 3D Visualization of space-time hot spots (Getis Ord) using Three.js
  • Andreas Wolf: Traffic foreasting with STARMA using Python
  • Manal Faraj: Traffic forecasting with LSTM Neural Networks (DL4J)
  • Salka Waterloo: Fixing missing data in time series using R
  • René Jablonski: Processing real-time traffic data using Apache Spark
  • Fabian Mallach: Route recommendation using a KNN approach
  • Yohan Burneleau: Interactive web app for taxi trips using Node.js and D3.js
  • Akshara Chukkannagari: Matching algorithm for WFS queries
  • Amgad Agoub: Storing GML application schemas in graph or multi model databases