Personalized Weighted Dijkstra Algorithm for Customized Path Recommendation

Le résumé fourni par la source

Customized path recommendation aims to provide personalized routing solutions that reflect individual user preferences, which is crucial for improving the travel experience across diverse user groups. Users’ path choices are influenced by a variety of factors, which are complex and interrelated. In this paper, we employ neural networks to model these intricate preferences and develop a customized path recommendation method. First, we extract users’ routing preferences from historical trajectory data. Then, we design neural networks to learn user preferences, with GRUs to predict sequential patterns. Finally, we propose a personalized weighted Dijkstra algorithm to generate customized paths for each user based on learned preferences. Case studies are conducted using a real trajectory data set. The results demonstrate that, compared to shortest path algorithms and frequency-based path selection methods, the proposed method can better fit users’ real path choices, showcasing its effectiveness in generating personalized routing solutions.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.