• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Transportation Research Circular no.E-C071; National Household Travel Survey Conference; 20041101-02; Washington, DC(US) >Data Fusion
【24h】

Data Fusion

机译：数据融合

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Data fusion is the process whereby two or more databases are integrated into a single source database, which is then used for statistical analysis. There are many policy, planning, and modeling issues for which National Household Transportation Survey (NHTS) data explain only part of the story, and data from other sources are necessary to complete the picture. Participants in this workshop identified opportunities for data fusion that uses NHTS data and examined the challenges and solutions for integrating one or more databases with NHTS data. Data fusion has been used in transportation research and modeling for many years. The associated processes and procedures are often expensive and labor intensive. Today, new opportunities exist to link various data sets together by using computer technologies, including emerging web applications and advances in geographic information systems software. Web applications are useful for fusing large data sets, such as the five files that make up the NHTS data set, into one super table. This merger makes it possible to perform cross-tabulation analysis with greater ease and flexibility. A common approach to data fusion appends data to individual sample households in the NHTS. For example, Claritas, a major marketer of consumer survey data, has developed lifestyle segments that are based on socioeconomic groups. Various characteristics, including percentage of owners and renters, percentage above or below median household income, and other important information, are fused to NHTS sample households. These fused characteristics are also being used in the transferability project, in which exact locations from the NHTS are matched to neighborhood cluster values, with each census tract being assigned to a cluster. It is basically a form of trip generation information.
机译：数据融合是将两个或多个数据库集成到单个源数据库中的过程，然后将其用于统计分析。对于许多政策，计划和建模问题，全国家庭运输调查（NHTS）数据仅能解释部分情况，而其他来源的数据也需要完整的图像。该研讨会的参与者发现了使用NHTS数据进行数据融合的机会，并探讨了将一个或多个数据库与NHTS数据集成的挑战和解决方案。数据融合已用于运输研究和建模多年。相关的过程和程序通常很昂贵且劳动强度大。如今，存在着使用计算机技术将各种数据集链接在一起的新机会，包括新兴的Web应用程序和地理信息系统软件的发展。 Web应用程序对于将大型数据集（例如组成NHTS数据集的五个文件）融合到一个超级表中很有用。通过合并，可以更轻松，更灵活地执行交叉表分析。数据融合的常用方法是将数据附加到NHTS中的各个样本住户。例如，消费者调查数据的主要营销者克拉丽塔斯（Claritas）根据社会经济群体开发了生活方式细分。 NHTS样本家庭融合了各种特征，包括所有者和租房者的百分比，家庭收入中位数以上或以下的百分比以及其他重要信息。这些融合的特征还用于可转移性项目中，其中将来自NHTS的确切位置与邻域聚类值匹配，并将每个普查区分配给一个聚类。它基本上是旅行生成信息的一种形式。

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