direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

Master Thesis: Collaborative Location Recommendations


Collaborative Location Recommendations


Over the past years, location-based services have been utilized in a wide range of mobile application domains, such as suggesting interesting places or routes based on the current location. Their operation provides enormous opportunities and ways to collect individuals’ whereabouts. Although there is a degree of randomness in a user’s mobility, there is a considerable lack of variability in their travel patterns. Several contributions in the field of location-based services and recommender systems have presented concepts and frameworks that can help in predicting places that a user might visit next. Predicting the immediate next place of a user is comparatively easy to achieve with acceptable accuracy, but the problem becomes more challenging when the prediction has to be made for the next day or week.

The goal of this thesis is to develop a concept and a prototypical implementation for a collaborative location recommendation system that can predict a set of places that are expected to be visited by a certain user on a certain day of the upcoming week. This thesis comprises of a compilation of background study, the development of a concept, prototypical implementation, and evaluation of the developed framework over a data set containing coarse-grained movement traces of users.

This thesis aims to predict a set of trajectories by observing the spatiotemporal periodicity of not only the associated user but also a set of other users who tend to follow similar spatiotemporal pattern over different days. The thesis uses various concepts including spatial clustering, temporal classification, spatiotemporal-periodic patterns extraction, trajectories construction, and sequential gap mining for making predictions. The evaluation of the developed frame- work over the provided data set infers that around 50% of trajectories were predicted with an average of 50% accuracy, i.e., half number of stays correctly predicted based on the geographical coordinates and the duration of the stay. This thesis is a contribution in the field of collaborative location recommendations that can find application in restricting epidemic outbreaks as well as urban planning by considering future user mobility.

Supervisor: Peter Ruppel

Type: Master Thesis

Duration: 6 months

Zusatzinformationen / Extras

Quick Access:

Schnellnavigation zur Seite über Nummerneingabe

Service-centric Networking
Telekom Innovation Laboratories
TEL 19
Ernst-Reuter-Platz 7
10587 Berlin, Germany
Phone: +49 30 8353 58811
Fax: +49 30 8353 58409