Two-Stage Tour Route Recommendation Approach by Integrating Crowd Dynamics Derived from Mobile Tracking Data

Tourism activities essentially represent the interaction between crowds and attractions. Thus, crowd dynamics are critical to quality of tourism experience in personalized tour recommendations. In order generate dynamic, routes, this paper develops a tourist trip design problem with (TTDP-CD), which is quantified indicators derived from mobile tracking data terms flow, interaction, structure. TTDP-CD attempts minimize perceived crowding maximize assessed value destinations while minimizing total distance proposes two-stage route strategy “global optimization first, local update later” deal sudden increase realistic scenarios. An evolutionary algorithm extended container-index coding, mixed mutation operators, global archive create day at urban scale. To corroborate performance approach, case study was carried out Dalian, China. The results demonstrate that suggested method outperforms previous approaches, such as NSGA-II, MOPSO, MOACO, WSM, solution decreases real-time by an average 7%.