Research On Passengers Travel Prediction And Recommendation Method

With the rapid development of mobile internet and wireless network technologies these years,more and more passengers in cities are relying on mobile apps to call taxicab to pick them up for travelling.This makes the knowledge about the potential passengers' requirements important and valuable.Therefore,how to use the historical trajectories to make prediction about people travelling becomes an important issue for the taxi transportation industry.In order to solve the problem mentioned above,this paper designs a multi-dimensional travelling conceptual model.Firstly,the relationship between the time,space and semantic of the three dimensions of passenger travel demand is defined,and the algorithm is designed to get spatio-temporal semantics from historical trajectories.Besides,the characters of passengers travelling is analyzed and the potential semantics of regional function is mined by using topic model.Then,based on the spatial clustering and time series analysis,the forecasting model is proposed,in which the parameters in time series analysis is determined by analysis of autocorrelation and partial autocorrelation.According to the problem of preferences in the spatial clustering,an adaptive algorithm is carried out to attain the goal.Next,the candidate set generated by a method combining hotness and Manhattan distance is recommended to taxicabs.Based on the real dataset provided by CAR INC.,the experiments shows that our approach gains a significant improvement in hotspot prediction and recommendation,with 15.21% improvement on average f-measure for prediction and 79.6% hit ratio for recommendation.In summary,this paper proposed an adaptive model which combines time-series forecasting techniques and spatio-temporal clustering method using historical taxi trajectories to predict passengers' demands in urban areas,then provided taxicab drivers with personalized recommendation.Actually,the prediction of hotspot not only help taxi driver find a passenger quickly,but also reduce the traffic jam.