东北大学学报（自然科学版）

东北大学学报:自然科学版 ›› 2020, Vol. 41 ›› Issue (6): 767-771.DOI: 10.12068/j.issn.1005-3026.2020.06.002

• 信息与控制 • 上一篇    下一篇

基于并行模式挖掘和路径匹配的用户位置预测

许贤泽， 谭盛煌， 刘静， 施元   

  1. (武汉大学 电子信息学院， 湖北 武汉430072)
  • 收稿日期:2019-07-29 修回日期:2019-07-29 出版日期:2020-06-15 发布日期:2020-06-12
  • 通讯作者: 许贤泽
  • 作者简介:许贤泽(1967-)，男，湖北京山人，武汉大学教授.
  • 基金资助:
    国家自然科学基金资助项目(51705375).

User Location Prediction Based on Parallel Pattern Mining and Path Matching

XU Xian-ze， TAN Sheng-huang， LIU Jing， SHI Yuan   

  1. Electronic Information School， Wuhan University， Wuhan 430072， China.
  • Received:2019-07-29 Revised:2019-07-29 Online:2020-06-15 Published:2020-06-12
  • Contact: TAN Sheng-huang
  • About author:-
  • Supported by:
    -

RichHTML

0

PDF (PC)

655

摘要： 为了提高移动用户位置预测的精度，提出了基于并行模式挖掘和路径匹配的移动用户位置预测方法，对传统的FP-GROWTH算法作了并行化处理，优化了节点负载分配方法，在Spark平台下挖掘用户移动频繁模式.改进了基于索引的路径相似度算法，提出基于路径最短距离的相斥度算法，提高了对轨迹数据缺失的适用性.在真实的用户轨迹数据集上实验表明，提出的基于轨迹相斥度预测方法相比马尔可夫模型和卡尔曼滤波模型拥有更高的预测精度，预测精确度平均提升7%左右.

关键词: 位置预测, Spark, FP-GROWTH, 模式挖掘, 轨迹相斥度

Abstract: In order to improve the accuracy of location prediction for mobile users， a method of location prediction for mobile users was proposed based on parallel pattern mining and path matching. Based on the traditional FP-GROWTH algorithm， the method of node load allocation was optimized， and frequent patterns of mobile users were mined on Spark platform. The index-based path similarity algorithm was improved， and the repulsion algorithm based on the shortest path distance was proposed to improve the applicability of missing trajectory data. Experiments on real user trajectory data sets show that the proposed model based on track dissimilarity prediction method has higher prediction accuracy than that of Markov model and Kalman filter model， which is improved by about 7% on average.

Key words: location prediction, Spark, FP-GROWTH, pattern mining, track dissimilarity

中图分类号: