Berechnung der Ähnlichkeit von Geographischen Objekten

• Applicant: Professorin Dr. Maike Buchin
• Subject Area: Theoretical Computer Science
• Term: from 2008 to 2010
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 81651471

We consider geographic objects such as the height values of a terrain, a sequence of positions stored by a GPS device, or a set of positions and times of a group of objects at certain events. These objects have a geometry and attributes which depend on space and possibly time. Due to recent advances in telecommunications, large amounts of geographic data is being acquired and stored in databases. From these, valuable information can be extracted using geographical analysis. The extracted information may help in decision making and prediction of future events. Because of the large amount of data, the analysis needs to be automated. For analysis tasks, such as the classification of geographic objects, the similarity of objects needs to be determined. This research aims at defining appropriate measures for the similarity of geographic objects and at designing efficient algorithms for computing these. The research is motivated by applications such as disease outbreak detection and classification of landforms. So far, little research has been conducted on similarity measures for geographic objects, whereas similarity measures for geometric objects have been wellresearched. For best results, we will combine techniques from geographical analysis and algorithm design, in particular computational geometry.

• DFG Programme: Research Fellowships
• International Connection: Netherlands
• Host: Professor Dr. Marc van Kreveld