英语翻译INFORMATION INTEGRATIONGiven the number of different statistical surveys and agencies conductingsurveys,“one-stop shopping” for federal statistical data wouldmake statistical data more accessible.Doing so depends on capabilitiesthat a

英语翻译INFORMATION INTEGRATIONGiven the number of different statistical surveys and agencies conductingsurveys,“one-stop shopping” for federal statistical data wouldmake statistical data more accessible.Doing so depends on capabilitiesthat a
英语翻译
INFORMATION INTEGRATION
Given the number of different statistical surveys and agencies conducting
surveys,“one-stop shopping” for federal statistical data would
make statistical data more accessible.Doing so depends on capabilities
that allow analyzing data from multiple sources.The goal would be to
facilitate both locating the relevant information across multiple surveys
and linking it to generate new results.Several possible approaches were
discussed at the workshop.
Metadata standards,including both standardized formats for describing
the data as well as sets of commonly agreed-on meanings,are one key
Summary of a Workshop on Information Technology Research for Federal Statistics
RESEARCH OPPORTUNITIES 31
to fully exploiting data sets from multiple sources.Without them,for
instance,it is very difficult to ascertain which fields in one data set correspond
to which fields in the other set and to what extent the fields are
comparable.While the framework provided by the recently developed
XML standard,including the associated data-type definitions (DTDs),
offers some degree of promise,work is needed to ensure that effective
DTDs for federal statistical data sets are defined.XML DTDs,because
they specify only certain structural characteristics of data,are only part of
the solution; approaches for defining the semantics of statistical data sets
also need to be developed.Standards do not,moreover,provide a solution
for legacy data sets.
Another approach to information integration is to leverage the existing
metadata,such as the text labels that describe the rows and columns
in a statistical table or descriptions of how the data have been collected
and processed,that accompany the data sets.Finding ways of using these
metadata to represent and relate the contents of tables and databases so
that analyses can be performed is an interesting area for further research.
The database community is exploring how to use database systems to
integrate information originating from different systems throughout an
organization (data warehousing).Database system developers are building
tools that provide an interactive,analytical front end that integrates
access to information in databases along with tools for visualizing the
data.Research is being done on such things as data transformations and
data cleaning and on how to model different data sources in an integrated
way.

英语翻译INFORMATION INTEGRATIONGiven the number of different statistical surveys and agencies conductingsurveys,“one-stop shopping” for federal statistical data wouldmake statistical data more accessible.Doing so depends on capabilitiesthat a
信息集成有多种不同的统计调查机构进行调查,"一站式购物"联邦统计数据,使统计数据更加方便.这样做取决于能力,使数据分析,从多种来源.目标是为了方便查找有关资料,在多个调查,并把它与创造新的 结果.几种可能的方法进行了讨论,在这次研讨会.数据标准,包括标准化的格式描述数据,以及套常用议定的含义,是一个关键研讨会纪要信息技术研究的联邦统计研究机会,31至充分利用 数据集,从多种来源.没有他们,比如 很难确定哪些领域,在一个数据集对应的领域,在其他设定 以及在何种程度上打球的人可比.而提供的框架最近开发的XML标准,包括相关的数据类型定义( DTDs ) 提供某种程度的承诺,禾 RK手术是为了确保有效的文档联邦统计数据集定义.XML文档,因为他们规定只有某些结构特征数据,都只能解决部分问题; 办法界定的语义学统计数据集,也需要得到发展.标准不,而且提供了新的解决遗留下来的数据集.另一个信息集成方法是利用现有的数据,如文字标签,说明行和列在统计表或说明如何 数据被收集和处理,即伴随数据集.如何利用这些数据来代表和涉及的内容,图表和数据库等,可以分析 演出将是一个有趣的地区作进一步研究.数据库社区正在探索如何利用数据库系统,以整合信息来自不同系统的一个有机组成 (数据仓库) .数据库系统开发商正在建设的工具,提供一个互动的,分析前端,将获取信息数据库随着工具扇区数据.科研方面正在做这样的事情,数据转换和数据清理,并就如何不同型号数据 源整合在一起.