• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>IEEE International Conference on Systems, Man and Cybernetics >A Grid-Based Clustering Method For Mining Frequent Trips From Large-Scale, Event-Based Telematics Datasets
【24h】

A Grid-Based Clustering Method For Mining Frequent Trips From Large-Scale, Event-Based Telematics Datasets

机译：一种基于网格的聚类方法，用于从大规模，基于事件的远程信息处理数据集进行频繁浏览

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Telematics systems that integrate wireless communications with sensor-based monitoring and location-aware applications have been widely deployed for mobile asset tracking and condition monitoring. In asset tracking field, exploring the data that relate to asset behaviors is critical to understand asset utilization, efficiency, distribution, operation, and many other important aspects in the supply chain. Prior work on analyzing GPS-based patterns has mainly been performed on time-based datasets. In this paper, we describe a scalable clustering algorithm to discover frequently repeated trips from large-scale, event-based telematics datasets collected via a satellite-based tracking system. We first transform GPS traces into a list of trips. Then we present a grid-based hierarchical clustering algorithm to discover frequent spatial patterns among all trips. We evaluate the effectiveness of the proposed algorithm against a large-scale, real-world dataset collected from tracking over a hundred of thousand assets and prove its feasibility. Through these experimental results, we show that the proposed algorithm significantly reduces the computational time needed for clustering as opposed to the traditional hierarchical clustering based on pair-wise comparison.
机译：为移动资产跟踪和条件监控集成了与基于传感器的监视和位置感知应用程序的无线通信的远程信息处理系统已被广泛部署。在资产跟踪领域，探索与资产行为相关的数据对于了解供应链中的资产利用，效率，分配，操作以及许多其他重要方面至关重要。在分析基于GPS的模式的情况下，主要是在基于时间的数据集上执行的。在本文中，我们描述了一种可扩展的聚类算法，以发现经常通过基于卫星的跟踪系统收集的大规模的基于事件的远程信息处理数据集的常用群集算法。我们首先将GPS跟踪转换为跳频列表。然后我们介绍基于网格的分层聚类算法，以发现所有旅行中的频繁的空间模式。我们评估所提出的算法对大规模的大型现实数据集的有效性，从追踪超过百万资产并证明其可行性。通过这些实验结果，我们表明，该算法显着降低了群集所需的计算时间，而基于配对比较，与传统的分层聚类相反。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号