Two-stage greedy algorithm based on crowd sensing for tour route recommendation

Abstract

Currently, the demand for tourism is increasing, but traditional tour group routes have been unable to meet individual needs. This paper proposes a novel personalized recommendation method for tour routes based on crowd sensing. First, we utilize ArcMap to model a real road network. We then propose a novel scoring mechanism for points of interest, including an interest label matching score and crowd sensing score, and implement a user-personalized multi-constraint interest model. Based on whether a user has must-see scenic spots, we propose a variable neighborhood greedy tour recommendation algorithm for users with no must-see scenic spots and a single-multiple point of interest two-stage greedy tour route recommendation algorithm for users with must-see scenic spots. We collected real data regarding 200 attractions, 881 restaurants, 570 hotels and 28 mature travel routes in Beijing from Ctrip, Dianping and Tuniu. We perform case analysis on Beijing dataset and comparative experiments on Beijing and public datasets with the existing algorithms. The experimental results demonstrate that our algorithm has superior rationality and efficiency.