Using GPS to learn significant locations and predict movement across multiple users
Personal and Ubiquitous Computing
Mining, indexing, and querying historical spatiotemporal data
Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining
A data mining approach for location prediction in mobile environments
Data & Knowledge Engineering
Extracting places from traces of locations
ACM SIGMOBILE Mobile Computing and Communications Review
Evaluating Next-Cell Predictors with Extensive Wi-Fi Mobility Data
IEEE Transactions on Mobile Computing
Mining interesting locations and travel sequences from GPS trajectories
Proceedings of the 18th international conference on World wide web
WhereNext: a location predictor on trajectory pattern mining
Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining
Trajectory simplification method for location-based social networking services
Proceedings of the 2009 International Workshop on Location Based Social Networks
Trip destination prediction based on past GPS log using a Hidden Markov Model
Expert Systems with Applications: An International Journal
Predestination: inferring destinations from partial trajectories
UbiComp'06 Proceedings of the 8th international conference on Ubiquitous Computing
Extracting significant places from mobile user GPS trajectories: a bearing change based approach
Proceedings of the 20th International Conference on Advances in Geographic Information Systems
Hi-index | 0.00 |
The positioning datasets that underpin route prediction models arrive as time series or point process logs. However, their use for prediction requires them to be split into meaningful segments, conceptualised as travelling periods or 'journeys', to form a set of training inputs. Despite significant research into route prediction, this important pre-processing step has traditionally occurred in an ad-hoc fashion, using arbitrary connectivity or movement thresholds. There has been little consideration to date of the impact of this on prediction, a fact rectified in this work. Three methods for detection of journeys are evaluated using a dataset of labelled movement histories collected specifically for this investigation. We perform an exhaustive series of parameterisations of detection methods, optimising with respect to actual journeys specified by users. We find that using existing methods, GPS multipath artefacts introduce journey extraction error that raises concern for prediction applications. A new approach is presented that explicitly uses these effects to improve journey detection results.