159
views
0
recommends
+1 Recommend
1 collections
    2
    shares
      • Record: found
      • Abstract: found
      • Article: found

      Trajectory Big Data Processing Based on Frequent Activity

      research-article

      Read this article at

      ScienceOpenPublisher
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          With the rapid development and wide use of Global Positioning System in technology tools, such as smart phones and touch pads, many people share their personal experience through their trajectories while visiting places of interest. Therefore, trajectory query processing has emerged in recent years to help users find their best trajectories. However, with the huge amount of trajectory points and text descriptions, such as the activities practiced by users at these points, organizing these data in the index becomes tedious. Therefore, the parallel method becomes indispensable. In this paper, we have investigated the problem of distributed trajectory query processing based on the distance and frequent activities. The query is specified by start and final points in the trajectory, the distance threshold, and a set of frequent activities involved in the point of interest of the trajectory. As a result, the query returns the shortest trajectory including the most frequent activities with high support and high confidence. To simplify the query processing, we have implemented the Distributed Mining Trajectory R-Tree index (DMTR-Tree). For this method, we initially managed the large trajectory dataset in distributed R-Tree indexes. Then, for each index, we applied the frequent itemset Apriori algorithm for each point to select the frequent activity set. For the faster computation of the above algorithms, we utilized the cluster computing framework of Apache Spark with MapReduce as the programing model. The experimental results show that the DMTR-Tree index and the query-processing algorithm are efficient and can achieve the scalability.

          Author and article information

          Journal
          Tsinghua Science and Technology
          Tsinghua Science and Technology
          Tsinghua University Press (Xueyan Building, Tsinghua University, Beijing 100084, China )
          1007-0214
          05 June 2019
          : 24
          : 3
          : 317-332
          Affiliations
          [1]∙ Amina Belhassena and Hongzhi Wang are with School of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001, China. E-mail: wangzh@hit.edu.cn.
          Author notes
          * To whom correspondence should be addressed. E-mail: amina_ belhasna@ 123456hotmail.fr ;

          Hongzhi Wang is a professor and doctoral supervisor at Harbin Institute of Technology. He received the PhD degree in computer science from Harbin Institute of Technology in 2018. He is a recipient of the outstanding dissertation award of CCF, Microsoft Fellow, and IBM PhD Fellowship. His research area is data management, including data quality, XML data management, and graph management. He has published more than 100 papers in refereed journals and conferences.

          Amina Belhassena received the PhD degree in computer science from Harbin Institute of Technology, China in 2018. She received the master degree of Technology in computer science from Abou bakr belkaid Tlemcen University, Algeria in 2012. Her research interest includes massive data computing, data mining, large-scale data management, and data indexing.

          Article
          1007-0214-24-3-317
          10.26599/TST.2018.9010087
          24d23902-bef4-40a8-8ed6-0f360d5a9b37
          Copyright @ 2019
          History
          : 22 November 2017
          : 22 February 2018

          Software engineering,Data structures & Algorithms,Applied computer science,Computer science,Artificial intelligence,Hardware architecture
          frequent activity,trajectory,distributed R-tree,query

          Comments

          Comment on this article