Inhalt des Dokuments
Es gibt keine deutsche Übersetzung dieser Webseite.
Master Thesis: The Analysis of Movement Traces for the Parametrization of Proactive Location-based Services
The Analysis of Movement Traces for the Parametrization of Proactive Location-based Services
In the recent years, location based services have become ubiquitous. From smartphone applications to connected cars, they are part of the everyday life of many people. Some of those services are intelligent: they provide information or trigger actions according to knowledge that they previously learned. They can gather a large amount of location data from devices, which can be analyzed to extract such knowledge. However, the size, the nature and the quality of the data poses new challenges for its analysis. Oneknowledgethatcanbeextractedfromthelocationdataofdevicesorusersisthefrequent movement patterns. A frequent movement pattern is a spatio-temporal sequence describing a movement that can be often observed in the dataset. This knowledge can be used to classify the movements and then used to provide adapted services to the end user. Multiple disciplines can beneﬁt from such knowledge, such as marketing, logistics, or trafﬁc engineering. In this thesis, the design and implementation of a framework to mine such frequent movement patterns is presented. The mined patterns are sequences of frequently visited spatial regions annotated with the frequent time intervals for devices to stay inside the regions and to transition between them. The evaluation of the framework shows that this system is able to retrieve a signiﬁcant number of frequent sequences from a dataset, and also that the optimizations brought in order to enable the framework to process relatively large datasets are viable.
Supervisor: Prof. Dr. Axel Küpper , Bersant
Deva, Sandro Rodriguez Garzon 
Type: Master Thesis
Duration: 6 months
10587 Berlin, Germany
Phone: +49 30 8353 58811
Fax: +49 30 8353 58409