Temporal-aware Location Prediction Model Using Similarity Approach

DOI

10.2991/3ca-13.2013.60How to use a DOI?

Keywords

Temporal-aware; Location Prediction; Trajectory Estimation; Movement Predictive Factors

Abstract

We propose scalable next location prediction model using temporal, movement behavior and trajectory history information. Most of the existing models use the whole information database and do not consider temporal information for location prediction; lead to scalability and accuracy issues. We consider road topology, multiple predictive factors and temporal information such as day and time to improve the accuracy and scalability issues. We extend the existing Mobility Markov Chain (MMC) model to incorporate n previously visited locations have movement factors mk in the time interval t of the day that we coined nmt-MMC. We use the whole information database in making the model in offline, and find similarity with the model for location and trajectory prediction in online. Simulation results demonstrate the trade-off between scalability and accuracy, and the effect of Top-N similar trajectories on the accuracy. We found that incorporating movement predictive factors improves the accuracy by approximately 7-percent, and adding both predictive factors and temporal information improve the accuracy by 14-percent.

Copyright

© 2013, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Ghulam Sarwar
AU  - Farman Ullah
AU  - Sungchang Lee
PY  - 2013/04
DA  - 2013/04
TI  - Temporal-aware Location Prediction Model Using Similarity Approach
BT  - Proceedings of the 2nd International Symposium on Computer, Communication, Control and Automation
PB  - Atlantis Press
SP  - 239
EP  - 243
SN  - 1951-6851
UR  - https://doi.org/10.2991/3ca-13.2013.60
DO  - 10.2991/3ca-13.2013.60
ID  - Sarwar2013/04
ER  -