Dunstan Matekenya
ABSTRACT
In location prediction systems, the purpose is to infer mobility pattern of individuals so that applications can know in advance where a user will go next. Although researchers have proposed many schemes to achieve better accuracy in location prediction systems, this area is still open to further research. In this study, we investigate the potential to improve performance (accuracy and training time) of location prediction models by leveraging large scale data. Given that users closer in space would exhibit similar mobility behaviors, our idea is to create what we are calling a community model for a group of users in a given geographic area and then use parameters from this model to enhance performance for individual users in the same community. We choose to experiment with logistic regression classifier and use a real life dataset to investigate this idea. The results from our experiments show that our idea to use community-wide learned model parameters in individuals works very well and reduces training time for individual models by nearly 100 percent. However, we don't find similar improvements in accuracy.
- Ashbrook, D., and Starner, T. Using gps to learn significant locations and predict movement across multiple users. Personal and Ubiquitous Computing 7, 5 (2003), 275--286. Google Scholar
Digital Library
- Etter, V., Kafsi, M., Kazemi, E., Grossglauser, M., and Thiran, P. Where to go from here? mobility prediction from instantaneous information. Pervasive and Mobile Computing 9, 6 (2013), 784--797. Mobile Data Challenge. Google ScholarDigital Library
- Friedman, J., Hastie, T., and Tibshirani, R. The elements of statistical learning, vol. 1. Springer series in statistics Springer, Berlin, 2001.Google Scholar
- Krumm, J., and Horvitz, E. Predestination: Inferring destinations from partial trajectories. In UbiComp 2006: Ubiquitous Computing. Springer, 2006, pp. 243--260. Google ScholarDigital Library
- McInerney, J., Rogers, A., and Jennings, N. R. Improving location prediction services for new users with probabilistic latent semantic analysis. In Proceedings of the 2012 ACM conference on ubiquitous computing (2012), ACM, pp. 906--910. Google ScholarDigital Library
- Song, L., Kotz, D., Jain, R., and He, X. Evaluating next-cell predictors with extensive wi-fi mobility data. Mobile Computing, IEEE Transactions on 5, 12 (2006), 1633--1649. Google ScholarDigital Library
Recommendations
Modeling dynamic spatiotemporal user preference for location prediction: a mutually enhanced method
AbstractAs the cornerstone of location-based services, location prediction aims to predict user’s next location through modeling user’s personal preference or travel sequential pattern. However, most existing methods only consider one of them and ...
Spatiotemporal prediction of foot traffic
LocalRec '21: Proceedings of the 5th ACM SIGSPATIAL International Workshop on Location-based Recommendations, Geosocial Networks and GeoadvertisingFoot traffic is a business term to describe the number of customers that enter a point of interest (POI). This work aims to predict future foot traffic: the number of people from each census block group (CBG) that will visit each POI of a study region ...
A data mining approach for location prediction in mobile environments
Mobility prediction is one of the most essential issues that need to be explored for mobility management in mobile computing systems. In this paper, we propose a new algorithm for predicting the next intercell movement of a mobile user in a Personal ...
Comments