Recommending Profitable Taxi Travel Routes Based on Big Taxi Trajectories Data

Recommending routes with the shortest cruising distance based on big taxi trajectories is an active research topic. In this paper, we first introduce a temporal probability grid network generated from the taxi trajectories, then a profitable route recommendation algorithm called Adaptive Shortest Expected Cruising Route (ASECR) algorithm is proposed. ASECR recommends profitable routes based on assigned potential profitable grids and updates the profitable route constantly based on taxis’ movements as well as utilizing the temporal probability grid network dynamically. To handle the big trajectory data and improve the efficiency of updating route constantly, a data structure kdS-tree is proposed and implemented for ASECR. The experiments on two real taxi trajectory datasets demonstrate the effectiveness and efficiency of the proposed algorithm.