Research On Travel Route Recommendation Technique Based On Hybrid Transportation Mode

With the increasingly improved of transportation network and the continuous increase of motor vehicles,route recommendation has drawn increasing attention recently as travel-ling is now an essential part of our daily life,especially for users who take taxis.However,existing route recommendation researches focus on single transportation mode,such as for the users who reach their destination by taxi only.This will undoubtedly cause a problem,that is,increasing traffic congestion can cause significant delays for taxi passengers and increase their travel time especially during the peak hours.Existing route recommendation researches based on single transportation mode cannot solve the problem of traffic congestion that lies in the very beginning and/or end of the trip.In addition,they did not evaluate the reliability of the recommendation.At the same time,public Bicycle Sharing System has been introduced in many cities around the world and bikers can pass through those congested roads more quickly than vehicles,which can effectively alleviate the start/last-mile problem.This means that on the same street,when there is heavy traffic,bikers may pass faster than taxis.To address the problem mentioned above,a Route Recommendation with Hybrid Transportation Modes model is designed,which aims to recommend the optimal route with the minimal travel time and corresponding transportation mode for a trip by taking taxis and/or shared bicycles.Besides,the thesis proposes a two-phase route recommendation model,which inte-grates a prediction and recommendation phase.The first phase proposes different learning models to predict the travel time by taxi/bicycle for a given trip and the station status in Bicy-cle Sharing System.In the route recommendation phase,the thesis develops a network-based travel time comparison algorithm to efficiently find the optimal route,where the algorithm combines three pruning strategies with high execution efficiency.Moreover,an evaluation criterion called Empirical Risk of Travel Time is designed to evaluate the reliability of our recommendation results.Then,the thesis verifies the effectiveness of proposed approaches through extensive experiments based on real datasets.Finally,the thesis has implemented a travel route recommendation system based on the hybrid traffic mode.For users' query of trip including the origin,destination,and departure time,the system can recommend the travel route with the least travel time,estimated travel time,the corresponding hybrid trans-portation mode,and the reliability evaluation of recommendation.