Profitable Taxi Travel Route Recommendation Based on Big Taxi Trajectory Data

Abstract

With the advent of GPS tracking technology, how to make use of taxi trajectories to efficiently and effectively reduce taxis cruising distance is an active and challenging research topic. In this paper, we propose a profitable taxi route recommendation method called adaptive shortest expected cruising route (ASER). In ASER, a probabilistic network model is developed to predict pick-up probability and capacity of each location by using Kalman filtering method. To recommend profitable driving routes to taxi drivers, ASER takes the load balance between passengers and taxis into consideration and the shortest expected cruising distance is introduced to formulate potential cruising distance of taxis. Moreover, MapReduce and a novel data structure kdS-tree are applied to improve recommendation efficiency. ASER is evaluated on two real trajectory datasets from San Francisco, CA, USA, and Wuhan, China. The experimental results validate that ASER significantly outperforms the existing methods by reducing the taxi cruising distance 11% and 39%.