Discovering Loose Group Movement Patterns from Animal Trajectories

The technical advances of positioning technologies enable us to track animal movements at finer spatial and temporal scales, and further help to discover a variety of complex interactive relationships. In this paper, considering the loose gathering characteristics of the real-life groups' members during the movements, we propose two kinds of loose group movement patterns and corresponding discovery algorithms. Firstly, we propose the weakly consistent group movement pattern which allows the gathering of a part of the members and individual temporary leave from the whole during the movements. To tolerate the high dispersion of the group at some moments (i.e. to adapt the discontinuity of the group's gatherings), we further scheme the weakly consistent and continuous group movement pattern. The extensive experimental analysis and comparison with the real and synthetic data shows that the group pattern discovery algorithms proposed in this paper are similar to the the real-life frequent divergences of the members during the movements, can discover more complete memberships, and have considerable performance.

[1]  Baoping Yan,et al.  Geographic variation in bar-headed geese Anser Indicus : connectivity of wintering areas and breeding grounds across a broad front , 2009 .

[2]  Htoo Htet Aung,et al.  Discovery of Evolving Convoys , 2010, SSDBM.

[3]  Xin Zhang,et al.  Fast mining of spatial collocations , 2004, KDD.

[4]  Panos Kalnis,et al.  On Discovering Moving Clusters in Spatio-temporal Data , 2005, SSTD.

[5]  Thomas Brinkhoff,et al.  A Framework for Generating Network-Based Moving Objects , 2002, GeoInformatica.

[6]  Joachim Gudmundsson,et al.  Reporting flock patterns , 2008, Comput. Geom..

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

[8]  Dino Pedreschi,et al.  Trajectory pattern mining , 2007, KDD '07.

[9]  Joachim Gudmundsson,et al.  Reporting Leaders and Followers among Trajectories of Moving Point Objects , 2008, GeoInformatica.

[10]  Nicholas Jing Yuan,et al.  On discovery of gathering patterns from trajectories , 2013, 2013 IEEE 29th International Conference on Data Engineering (ICDE).

[11]  Jing Yuan,et al.  A framework of traveling companion discovery on trajectory data streams , 2013, ACM Trans. Intell. Syst. Technol..

[12]  Jae-Gil Lee,et al.  MoveMine: Mining moving object data for discovery of animal movement patterns , 2011, TIST.

[13]  Joachim Gudmundsson,et al.  Computing longest duration flocks in trajectory data , 2006, GIS '06.

[14]  Patrick Laube,et al.  Analyzing Relative Motion within Groups of Trackable Moving Point Objects , 2002, GIScience.

[15]  Baoping Yan,et al.  Mining Common Spatial-Temporal Periodic Patterns of Animal Movement , 2013, 2013 IEEE 9th International Conference on e-Science.

[16]  Christian S. Jensen,et al.  Discovery of convoys in trajectory databases , 2008, Proc. VLDB Endow..

[17]  Jiawei Han,et al.  Swarm: Mining Relaxed Temporal Moving Object Clusters , 2010, Proc. VLDB Endow..

[18]  Baoping Yan,et al.  Wild Bird Migration across the Qinghai-Tibetan Plateau: A Transmission Route for Highly Pathogenic H5N1 , 2011, PloS one.