Abstract
The ranking and recommendation of regions of interest are increasingly important in recent years. In this light, we propose and study a novel and interesting problem of inferring region significance using multi-source spatiotemporal data. In our study, POIs, locations, regions, trajectories, and spatial networks are taken into account. Given a set of regions R and a set of trajectories T, we seek for the top-k most attractive regions to users, i.e., regions with the top-k highest spatial-density correlations to the trajectories of travelers. This study is useful in many mobile applications such as urban computing, region recommendation, and location-based service in general. This problem is challenging due to two reasons: (1) how to model the spatial-density correlation effectively and practically and (2) how to process the problem in interactive time. To overcome the challenges, we design a novel spatial-density correlation function to evaluate the relationship between regions and trajectories, and the density of POIs and network distance are taken into account. Then, we develop a series of optimization techniques to accelerate the query efficiency. Furthermore, we develop a parallel mechanism to support big spatial data. Finally, we conduct extensive experiments on real and synthetic spatial data sets to show the efficiency and effectiveness of developed algorithms.
Paper version not known (
Free)
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call