On Feeding Business Systems with Linked Resources from the Web of Data

Abstract

Business systems that are fed with data from the Web of Data require transparent interoperability. The Linked Data principles establish that different resources that represent the same real-world entities must be linked for such purpose. Link rules are paramount to transparent interoperability since they produce the links between resources. State-of-the-art link rules are learnt by genetic programming and build on comparing the values of the attributes of the resources. Unfortunately, this approach falls short in cases in which resources have similar values for their attributes, but represent different real-world entities. In this paper, we present a proposal that leverages a genetic programming that learns link rules and an ad-hoc filtering technique that boosts them to decide whether the links that they produce must be selected or not. Our analysis of the literature reveals that our approach is novel and our experimental analysis confirms that it helps improve the F1 score by increasing precision without a significant penalty on recall.Supported by the Spanish R&D programme (grants TIN2013-40848-R and TIN2016-75394-R). The computing facilities were provided by the Andalusian Scientific Computing Centre (CICA). We are grateful to Dr. Carlos R. Rivero and Dr. David Ruiz for earlier ideas that led to the results in this paper. We also thank Dr. Francisco Herrera for his hints on statistical analyses and sharing his software with us.

Helio: A framework for implementing the life cycle of knowledge graphs

“…Enriching covers a large number of possible tasks [69], from transforming RDF data in order to increase its quality, e.g., removing white spaces or capitalising names, up to creating new data on the fly, e.g., completing the RDF data of a KG with machine learning [1]. Linking aims at producing links between the RDF resources by means of link rules [15]. Additionally, these links may involve RDF resources from different KGs; creating these links is one of the Linked Data principles [3].…”

“…RDF link discovery aims at producing relationships between local RDF resources and other RDF resources allocated in different KGs [58]. On the one hand, there is a wide number of tools that aim at producing link rules [16][17][18]23,43,44,60,61,75], i.e., restrictions under which two RDF resources are linked. On the other hand, other tools focus on applying those rules efficiently and producing the links among resources [26,59,79].…”

“…The RDF Generator Module has several internal translators in order to understand different mapping languages, like RML, 15 the WoT-Mappings, or the JSON serialisation 16 of the model depicted by Fig. 4.…”

“…For this reason, the RDF Generator Module counts with a dynamic system for loading plugins. 17 This allows users to develop new data providers or handlers without modifying the code of Helio, and load these extensions dynamically.…”

On extracting data from tables that are encoded using HTML

“…Before feeding the record sets returned by data extraction into a particular application, it is commonly necessary to perform some of the following integration tasks: semantisation [25,45,54,55,60,63,71], which either maps the descriptors onto the terminology box of a particular ontology or the tuples onto its assertion box [19]; union [23], which merges record sets that provide similar data; finding primary keys [62], which determines which components of the tuples identify them as univocally as possible; record linkage [8,11,12], which finds different records that refer to the same actual entities; augmentation [6,52,67], which joins record sets on the same topic to complete the information that they provide individually; and cleaning [10,31,61], which fixes data. Note that the integration tasks are orthogonal to data extraction because they are independent from the source of the record sets, which is the reason why they fall out of the scope of this article.…”