Travel purpose identification based on mobile phone signaling data and Bayesian network

It is an important issue in the field of transportation planning to identify and detect the travel purpose of individual travelers. The mobile phone signaling data has the advantages of low collection cost, wide user coverage, and strong real-time performance in the 4G/5G networking environment. However, it is difficult to mine the travel characteristics effectively from the mobile phone signaling data with a large proportion of noise data to identify the travel purpose. This paper relies on the mobile phone signaling data of residents in Kunshan City, Jiangsu Province, China, to obtain individual travel segments, and then combines residents' travel survey data and POI data to mine individual travel information. The Bayesian network is preliminarily constructed through the constraint-based Bayesian network structure learning theory. This article takes the travel purpose as the deductive reasoning object, prunes the Bayesian network and achieves travel purpose identification. The results show that, based on the travel feature mining method proposed in this paper, the accuracy of travel purpose identification can reach 91.28% through the Bayesian network identification model.