RT Conference Proceedings
T1 Analysis of location prediction performance of LZ algorithms using GSM Cell-based location data
A1 Rodriguez-Carrión, Alicia
A1 García Rubio, Carlos
A1 Campo Vázquez, María Celeste
A1 Cortés Martín, Alberto
A1 García Lozano, Estrella María
A1 Noriega-Vivas, Patricia
AB Predictions about users' next locations allowbringing forward their future context, thus having additionaltime to react. To make such predictions, algorithms capable oflearning mobility patterns and estimating the next location areneeded. This work is focused on making the predictions onmobile terminals, thus resource consumption being an important constraint. Among the predictors with low resourceconsumption, the family of LZ algorithms has been chosen to study their performance, analyzing the results drawn fromprocessing location records of 95 users. The main contribution is to divide the algorithms into two phases, thus being possible to use the best combination to obtain better prediction accuracy or lower resource consumption.
PB UCAMI
YR 2011
FD 2011
LK https://hdl.handle.net/10016/13103
UL https://hdl.handle.net/10016/13103
LA eng
NO Proceedings of the 5th International Symposium of Ubiquitous Computing and Ambient Intelligence (UCAMI 2011), December 5-8th, 2011, Riviera Maya, Mexico
NO Proyecto CCG10-UC3M/TIC-4992 de la Comunidad Autónoma de Madrid y la Universidad Carlos III de Madrid
DS e-Archivo
RD 27 jul. 2024