Mobile data mining based on trajectories clustering

作者：Tang, Zhi-Hang; Li, Rong-Jun

来源：International Review on Computers and Software, 2012, 7(2): 601-605.

• Clustering accuracy
• Hilbert curve
• Mobile data mining
• Similarity measurements
• Spatial partitioning
• Spatial regions
• Trajectory points

To address the problem of uncertain trajectories, we suggest a candidate segments approach to uncertain clustering accuracy. This approach is based on a Hilbert curve spatial partitioning to map the trajectory points to spatial regions. Based on candidate regions of uncertain trajectories segments, we construct candidate segments to model uncertain trajectories. For the large number of candidate segments, we use a sketch-based approach in order to create hash-compressed candidate segments representations of the clusters. Based on the sketch, we show a similarity measurement algorithm of incoming uncertain trajectory to clusters. An algorithm for determining the assignment of incoming trajectory to clusters is given. Final, we verify the accuracy and efficiency of the proposed scheme via experiments.

• 出版日期2012
• 单位华南理工大学