SCHISM: A scheme for identification of suspicious mobile clusters for location-based services

Location-based services (LBS) deliver location-dependent content to mobile users. Mobile users are equipped with devices, e.g., smartphones, which regularly report the position to a back-end system (BES). Users, with similar moving patterns, might also have requests for similar content delivery (e.g., museum/city-tour guidance, traffic conditions). Moreover, users can be formed into (temporal) groups, whose structure can vary over time (i.e., group-merge/split, group membership). The larger the number of users is, the higher the overall BES load becomes. In this paper, we propose a group management scheme (SCHISM) which exploits the users' formation into groups in order to tackle the BES-mobile device communication load. Specifically, we treat a group as a single user (the group leader; GL). In this case, the BES communicates only with the GL for certain time horizon. The GL can then disseminate the LBS content to the group members. We adopt an agglomerative clustering scheme for the group formation phase. The clustering gain is sequentially monitored corresponding to the transitions between consecutive merge-steps. A user group is classified as `suspicious', i.e., candidate for fragmentation/split, when a decrease in the clustering gain is obtained during the group formation phase. Performance assessment reveals significant improvements due to the considered scheme for location-based systems.

[1]  Hans-Peter Kriegel,et al.  A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise , 1996, KDD.

[2]  Ding-Zhu Du,et al.  A Decision Criterion for the Optimal Number of Clusters in Hierarchical Clustering , 2003, J. Glob. Optim..

[3]  Hamid R. Rabiee,et al.  MobiSim: A Framework for Simulation of Mobility Models in Mobile Ad-Hoc Networks , 2007 .

[4]  Hans-Peter Kriegel,et al.  Clustering Moving Objects via Medoid Clusterings , 2005, SSDBM.

[5]  Peter J. Rousseeuw,et al.  Finding Groups in Data: An Introduction to Cluster Analysis , 1990 .

[6]  Charles Elkan,et al.  Scalability for clustering algorithms revisited , 2000, SKDD.

[7]  Anil K. Jain,et al.  Data clustering: a review , 1999, CSUR.

[8]  Tian Zhang,et al.  BIRCH: an efficient data clustering method for very large databases , 1996, SIGMOD '96.

[9]  J. H. Ward.  Hierarchical Grouping to Optimize an Objective Function , 1963 .

[10]  Beng Chin Ooi,et al.  Continuous Clustering of Moving Objects , 2007, IEEE Transactions on Knowledge and Data Engineering.

[11]  Injong Rhee,et al.  On the levy-walk nature of human mobility , 2011, TNET.

[12]  Sudipto Guha,et al.  CURE: an efficient clustering algorithm for large databases , 1998, SIGMOD '98.

[13]  Yannis Manolopoulos,et al.  C2P: Clustering based on Closest Pairs , 2001, VLDB.

[14]  Anil K. Jain,et al.  Algorithms for Clustering Data , 1988 .

[15]  Hans-Peter Kriegel,et al.  OPTICS: ordering points to identify the clustering structure , 1999, SIGMOD '99.

[16]  Xiaoyan Hong,et al.  A group mobility model for ad hoc wireless networks , 1999, MSWiM '99.