2017
DOI: 10.1007/978-3-319-58068-5_5
|View full text |Cite
|
Sign up to set email alerts
|

Methods for Intrinsic Evaluation of Links in the Web of Data

Abstract: Abstract. The current Web of Data contains a large amount of interlinked data. However, there is still a limited understanding about the quality of the links connecting entities of different and distributed data sets. Our goal is to provide a collection of indicators that help assess existing interlinking. In this paper, we present a framework for the intrinsic evaluation of RDF links, based on core principles of Web data integration and foundations of Information Retrieval. We measure the extent to which link… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
6
0

Year Published

2017
2017
2024
2024

Publication Types

Select...
4
2
2

Relationship

0
8

Authors

Journals

citations
Cited by 8 publications
(6 citation statements)
references
References 13 publications
0
6
0
Order By: Relevance
“…Measuring this information, and specifically how much information is gained by connecting two knowledge graphs together is a major aspect of this work. Here, we were strongly influenced by the work of Sarasua et al [19], who investigate the value of links between different data sets of a knowledge graph. To do so, statistics (based on [1]), such as the net amount of links, the average number of incoming and outgoing links per data set and number of entities linked by a source, are calculated.…”
Section: Related Workmentioning
confidence: 99%
See 1 more Smart Citation
“…Measuring this information, and specifically how much information is gained by connecting two knowledge graphs together is a major aspect of this work. Here, we were strongly influenced by the work of Sarasua et al [19], who investigate the value of links between different data sets of a knowledge graph. To do so, statistics (based on [1]), such as the net amount of links, the average number of incoming and outgoing links per data set and number of entities linked by a source, are calculated.…”
Section: Related Workmentioning
confidence: 99%
“…In the second step, the intersection of the graphs is determined. If the parties still want to continue after this step, a set of metrics (based on [19]) is computed, while still maintaining the graphs secret. The outcome of the measures should give the Buyer an indication on whether the graph is interesting, but to determine the quality, actual data needs to be transferred (which could involve monetary exchange).…”
Section: Approachmentioning
confidence: 99%
“…Such algorithms may or may not exploit ontological information, but this does not affect our central hypothesis. [14] also uses network structure to evaluate link quality, but in a very different way. The main intuition there is that an individual link in an ILN is more reliable when it leads to a greater information gain.…”
Section: Related Workmentioning
confidence: 99%
“…Information Gain. The work in [14] also uses network structure to evaluate link quality, but in a very different way. The main intuition there is that an individual link in an ILN is more reliable when it leads to a greater information gain.…”
Section: Related Workmentioning
confidence: 99%