Contextual Subgraph Discovery with Mobility Models

Starting from a relational database that gathers information on people mobility - such as origin/destination places, date and time, means of transport - as well as demographic data, we adopt a graph-based representation that results from the aggregation of individual travels. In such a graph, the vertices are places or points of interest (POI) and the edges stand for the trips. Travel information as well as user demographics are labels associated to the edges. We tackle the problem of discovering exceptional contextual subgraphs, i.e., subgraphs related to a context - a restriction on the attribute values - that are unexpected according to a model. Previous work considers a simple model based on the number of trips associated with an edge without taking into account its length or the surrounding demography. In this article, we consider richer models based on statistical physics and demonstrate their ability to capture complex phenomena which were previously ignored.

[1]  Jae-Gil Lee,et al.  MoveMine: mining moving object databases , 2010, SIGMOD Conference.

[2]  Martin Atzmüller,et al.  Detecting community patterns capturing exceptional link trails , 2016, 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM).

[3]  Thad Starner,et al.  Using GPS to learn significant locations and predict movement across multiple users , 2003, Personal and Ubiquitous Computing.

[4]  Gerardo Chowell,et al.  Null Models for Community Detection in Spatially-Embedded, Temporal Networks , 2014, bioRxiv.

[5]  Marta C. González,et al.  A universal model for mobility and migration patterns , 2011, Nature.

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

[7]  Andreas Hotho,et al.  Mining Subgroups with Exceptional Transition Behavior , 2016, KDD.

[8]  Xing Xie,et al.  Mining interesting locations and travel sequences from GPS trajectories , 2009, WWW '09.

[9]  Renaud Lambiotte,et al.  Uncovering space-independent communities in spatial networks , 2010, Proceedings of the National Academy of Sciences.

[10]  Atri Rudra,et al.  iMAP: Indirect measurement of air pollution with cellphones , 2009, 2009 IEEE International Conference on Pervasive Computing and Communications.

[11]  Peter A. Flach,et al.  Rule Evaluation Measures: A Unifying View , 1999, ILP.

[12]  Vinny Cahill,et al.  Route profiling: putting context to work , 2004, SAC '04.

[13]  P. Borgnat,et al.  Enhancing Space-Aware Community Detection Using Degree Constrained Spatial Null Model , 2017 .

[14]  M. Batty,et al.  Gravity versus radiation models: on the importance of scale and heterogeneity in commuting flows. , 2012, Physical review. E, Statistical, nonlinear, and soft matter physics.

[15]  Céline Robardet,et al.  Exceptional contextual subgraph mining , 2017, Machine Learning.

[16]  The P 1 P 2/D Hypothesis: The Case of Railway Express , 1946 .

[17]  Arno Knobbe,et al.  Exceptional Model Mining , 2008, ECML/PKDD.

[18]  Lei Chen,et al.  Finding time period-based most frequent path in big trajectory data , 2013, SIGMOD '13.

[19]  Murat Ali Bayir,et al.  Track me! a web based location tracking and analysis system for smart phone users , 2009, 2009 24th International Symposium on Computer and Information Sciences.

[20]  Dino Pedreschi,et al.  Human mobility, social ties, and link prediction , 2011, KDD.

[21]  M. Nanni.  Mobility , Data Mining and Privacy – the GeoPKDD project , 2009 .

[22]  Joseph S. Fulda.  Data Mining and Privacy , 2000 .