Next place prediction by understanding mobility patterns

As technology to connect people across the world is advancing, there should be corresponding advancement in taking advantage of data that is generated out of such connection. To that end, next place prediction is an important problem for mobility data. In this paper we propose several models using dynamic Bayesian network (DBN). Idea behind development of these models come from typical daily mobility patterns a user have. Three features (location, day of the week (DoW), and time of the day (ToD)) and their combinations are used to develop these models. Knowing that not all models work well for all situations, we developed three combined models using least entropy, highest probability and ensemble. Extensive performance study is conducted to compare these models over two different mobility data sets: a CDR data and Nokia mobile data which is based on GPS. Results show that least entropy and highest probability DBNs perform the best.

[1]  Albert-László Barabási,et al.  Limits of Predictability in Human Mobility , 2010, Science.

[2]  Qiang Yang,et al.  Feature engineering for semantic place prediction , 2013, Pervasive Mob. Comput..

[3]  Bruno Martins,et al.  Predicting future locations with hidden Markov models , 2012, UbiComp.

[4]  Geoffrey Zweig,et al.  Speech Recognition with Dynamic Bayesian Networks , 1998, AAAI/IAAI.

[5]  Peter J. F. Lucas,et al.  Dynamic Bayesian networks as prognostic models for clinical patient management , 2008, J. Biomed. Informatics.

[6]  Dieter Fox,et al.  Bayesian Filtering for Location Estimation , 2003, IEEE Pervasive Comput..

[7]  Kun Chang Lee,et al.  A Dynamic Bayesian Network Approach to Location Prediction in Ubiquitous Computing Environments , 2010, IAIT.

[8]  Cecilia Mascolo,et al.  Mining User Mobility Features for Next Place Prediction in Location-Based Services , 2012, 2012 IEEE 12th International Conference on Data Mining.

[9]  Jingjing Wang,et al.  Periodicity Based Next Place Prediction , 2012 .

[10]  Ilse Ravyse,et al.  DBN Based Models for Audio-Visual Speech Analysis and Recognition , 2006, PCM.

[11]  K. Allan.  Classifiers , 2015 .

[12]  Kevin P. Murphy,et al.  Learning the Structure of Dynamic Probabilistic Networks , 1998, UAI.

[13]  Jorge Dias,et al.  Gesture Recognition Using a Marionette Model and Dynamic Bayesian Networks (DBNs) , 2006, ICIAR.

[14]  Fabio Porto,et al.  A conceptual view on trajectories , 2008, Data Knowl. Eng..

[15]  Xin Lu,et al.  Approaching the Limit of Predictability in Human Mobility , 2013, Scientific Reports.

[16]  Imad Aad,et al.  The Mobile Data Challenge: Big Data for Mobile Computing Research , 2012 .

[17]  Ehsan Kazemi,et al.  Been There, Done That: What Your Mobility Traces Reveal about Your Behavior , 2012 .

[18]  Yang Xiang,et al.  Multiagent bayesian forecasting of structural time-invariant dynamic systems with graphical models , 2011, Int. J. Approx. Reason..

[19]  Ian Oakley,et al.  Indoor-ALPS: an adaptive indoor location prediction system , 2014, UbiComp.

[20]  Anna Monreale,et al.  WhereNext: a location predictor on trajectory pattern mining , 2009, KDD.

[21]  Wei Wu,et al.  Oscillation Resolution for Mobile Phone Cellular Tower Data to Enable Mobility Modelling , 2014, 2014 IEEE 15th International Conference on Mobile Data Management.

[22]  Daniel Gatica-Perez,et al.  Contextual conditional models for smartphone-based human mobility prediction , 2012, UbiComp.

[23]  Qiang Yang,et al.  Domain-constrained semi-supervised mining of tracking models in sensor networks , 2007, KDD '07.