A Novel Personalized Dynamic Route Recommendation Approach Based on Pearson Similarity Coefficient in Cooperative Vehicle-Infrastructure Systems

This paper mainly proposes the theoretical model and practical method of a novel personalized dynamic route recommendation system under Cooperative Vehicle-Infrastructure System. Considering the complexity of travelers preference factors and the time dependence of route feature vector, we introduce Pearson similarity coefficient to measure the correlation between the travelers preference feature vector and the route feature vector, and the route with the highest similarity is recommended to the traveler. Meanwhile, the quantitative analysis and modeling of the corresponding factors are realized for the negative factors in reality such as weather, accidents and temporary traffic control. In particular, this work also designs the Feasible Route Search Algorithm, which could search all feasible routes between the origins and the destinations. In order to verify the feasibility of the proposed route recommendation model, we apply Baidu map system development tools to develop a dynamic route recommendation simulation system with perfect function and user-friendly interface by using C# language. The system displays various functions in the way of simulating real-time navigation.