基于数据挖掘的移动用户出行轨迹预测

Travel Trajectory Prediction of Mobile Users Based on Data Mining

为了解决移动用户出行轨迹预测的问题,首先利用用户出行轨迹数据进行语义化建模,然后根据语义位置和访问概率对用户群进行分类,再次,利用关联规则挖掘不同群体的频繁模式,最后,结合实时出行数据动态更新贝叶斯网络实现移动用户出行轨迹的实时预测。经过实验表明,该算法能够在一定程度上反映用户出行的目的和偏好,并具有很好的扩展性。

In order to solve the problem of trajectory prediction for mobile users, this paper firstly uses trajectory data for semantic modeling, and then classifies user groups according to their semantic location and access probability. The association rules are used to mine frequent patterns of different groups. Finally, the Bayesian network is updated dynamically with real-time travel data to realize real-time travel trajectory prediction for mobile users. Experiments show that the algorithm can reflect users ’travel purposes and preferences to a certain extent, and has good scalability.