Graph-Based Approach for Personalized Travel Recommendations

In the evolving domain of urban mobility systems, the integration of technology with user-centric strategies is pivotal. This research stands on the foundational concept of Mobility-as-a-Service, a user-centric intelligent mobility management distribution system that seeks to prioritize human needs over mere technological infrastructure. The study delves deep into the wealth of data available through mobile sensing technologies, highlighting the unprecedented understanding it offers into human mobility patterns, thus facilitating personalized route recommendations. The literature categorizes the study area into three interlinked categories: point-of-interest (POI) recommendation, travel planning, and trajectory modelling. In a significant stride, this research introduces a comprehensive understanding of hu-man mobility data and proposes a novel framework designed to tender personalized recommendations to travel planner users. The innovative framework employs a graph-based approach rooted in the Sussex-Huawei Locomotion dataset, leveraging an adaptation of Bellman-Ford’s algorithm. This modification considers factors such as perceived fatigue, frequency of trips to specific locations, and proximity to POIs, promising a path routed in past user experiences and preferences.