A hybrid grid-based method for mining arbitrary regions-of-interest from trajectories

There is an increasing need for a trajectory pattern mining as the volume of available trajectory data grows at an unprecedented rate with the aid of mobile sensing. Region-of-interest mining identifies interesting hot spots that reveal trajectory concentrations. This article introduces an efficient and effective grid-based region-of-interest mining method that is linear to the number of grid cells, and is able to detect arbitrary shapes of regions-of-interest. The proposed algorithm is robust and applicable to continuous and discrete trajectories, and relatively insensitive to parameter values. Experiments show promising results which demonstrate benefits of the proposed algorithm.

N/A

Workshop on Machine Learning for Sensory Data Analysis

N/A

978-1-4503-2513-4

N/A

N/A

8