An evaluation of uplift mapping languages
An evaluation of uplift mapping languages
复制标题
DOI：
10.1108/ijwis-04-2017-0036
复制
发表时间：
2017-01-01
影响因子：
1.6
通讯作者：
O'Sullivan, Declan
中科院分区：
文献类型：
--
作者：
Crotti, Ademar, Jr.;Debruyne, Christophe;O'Sullivan, Declan
目的-本文旨在评估CSV提升工具的最新技术水平。在此基础上，提出并评价了一种通过函数将数据转换成提升映射语言的方法。通常，当需要数据转换时，将非资源描述框架（RDF）数据映射为RDF格式的工具依赖于数据源本身的技术。根据数据格式，可以使用底层技术执行数据操作，例如用于关系数据库的关系数据库管理系统（RDBMS）或用于XML的XML。对于CSV/表格数据，没有这样的底层技术，而是需要将源数据转换为另一种格式或预/后处理技术。设计/方法/方法-为了评估CSV提升工具的最新技术，作者提出了一个比较框架，并将其应用于此类工具。在比较框架中评估的一个关键特性是数据转换功能。他们认为，现有的转换功能的方法是复杂的-因为需要一些步骤和工具。所提出的方法，FunUL，相比之下，定义的功能独立的源数据映射到RDF中，作为资源内的mapping itself.Findings -该方法进行了评估，使用两个典型的现实世界的用例。作者比较了我们的方法和其他方法（包括转换函数作为提升映射的一部分）如何实现从CSV/Tabular到RDF的提升映射。这种比较表明，作者的方法表现良好，这些用例。独创性/价值-本文提出了一个比较框架，并将其应用到国家的最先进的CSV提升工具。此外，作者描述了FunUL，与其他相关工作不同，它将功能定义为提升映射本身中的资源，集成了数据转换功能和映射定义。这使得从源数据生成RDF变得透明和可追踪。此外，由于函数被定义为资源，这些资源可以在映射中多次重用。
Purpose - This paper aims to evaluate the state-of-the-art in CSV uplift tools. Based on this evaluation, a method that incorporates data transformations into uplift mapping languages by means of functions is proposed and evaluated. Typically, tools that map non-resource description framework (RDF) data into RDF format rely on the technology native to the source of the data when data transformation is required. Depending on the data format, data manipulation can be performed using underlying technology, such as relational database management system (RDBMS) for relational databases or XPath for XML. For CSV/Tabular data, there is no such underlying technology, and instead, it requires either a transformation of source data into another format or pre/post-processing techniques.Design/methodology/approach - To evaluate the state-of-the-art in CSV uplift tools, the authors present a comparison framework and have applied it to such tools. A key feature evaluated in the comparison framework is data transformation functions. They argue that existing approaches for transformation functions are complex - in that a number of steps and tools are required. The proposed method, FunUL, in contrast, defines functions independent of the source data being mapped into RDF, as resources within the mapping itself.Findings - The approach was evaluated using two typical real-world use cases. The authors have compared how well our approach and others (that include transformation functions as part of the uplift mapping) could implement an uplift mapping from CSV/Tabular into RDF. This comparison indicates that the authors' approach performs well for these use cases.Originality/value - This paper presents a comparison framework and applies it to the state-of-the-art in CSV uplift tools. Furthermore, the authors describe FunUL, which, unlike other related work, defines functions as resources within the uplift mapping itself, integrating data transformation functions and mapping definitions. This makes the generation of RDF from source data transparent and traceable. Moreover, as functions are defined as resources, these can be reused multiple times within mappings.