Travel Behavior Pattern Mining Based On Cellular Data

With the rapid development of mobile networks,cellular data is more easily to obtain.Compared to GPS data,cellular data has a larger coverage of users and a longer time span.Besides,the scale of cellular data is easy to expand,which is the advantage of cellular data set.The tourist behavior mining is based on cellular data provided by operator.Great efforts has been made on the classification of tourists and the identification of group tourists.A method of identifying group tourists has been designed.The innovation and main work of this paper is embodied in the following aspects:First,the design and implementation of tourist hierarchy classification framework,which including the module of data preprocessing and data analyzing.Second,a group identification algorithm based on similarity is proposed in this work.The algorithm is suitable for group identification on low precision cellular data,and the user's similarity measurement is introduced to judge the user's peer relationship.Third,traffic pattern recognition obtains the travel methods of tourists through the trajectories of tourists' and map information.This work also verifies the accuracy of results of group identification from the side.The method we used has made innovations on the existing trajectory data mining methods,and the results of tourists behavior analysis have reference value for tourism industry.