What Should I Link to? Identifying Relevant Sources and Classes for Data Linking

Abstract

Abstract. With more data repositories constantly being published on the Web, choosing appropriate data sources to interlink with newly published datasets becomes a non-trivial problem. It is necessary to choose both the repositories to link to and the relevant subsets of these repositories, which contain potentially matching individuals. In order to do this, detailed information about the content and structure of semantic repositories is often required. However, retrieving and processing such information for a potentially large number of datasets is practically unfeasible. In this paper, we propose an approach which utilises an existing semantic web index in order to identify potentially relevant datasets for interlinking and rank them. Furthermore, we adapt instance-based ontology schema matching to extract relevant subsets of selected data source and, in this way, pre-configure data linking tools.

Two Approaches to the Dataset Interlinking Recommendation Problem

“…Nikolov et al [4,5] propose an approach to identify relevant datasets for interlinking, with two main steps: (i) searching for potentially relevant entities in other datasets using as keywords a subset of labels in the new published dataset; and (ii) filtering out irrelevant datasets by measuring concept similarities obtained by applying ontology matching techniques.…”

“…As features of lak, we used three classes, swc:ConferenceEvent, swrc:Proceedings and swrc:InProceedings, obtained from the LinkedUp project Web site 5 . As the candidates to be ranked, we selected the datasets webscience, webconf, wordnet, dblp and courseware.…”

Recommending Tripleset Interlinking through a Social Network Approach

“…Nikolov et al [3,16] propose an approach to identify relevant triplesets for data linking. Their approach establishes two main steps: (i) searching for potentially relevant resources in other triplesets using as keywords a subset of labels in the new published tripleset; and (ii) filtering out irrelevant triplesets by measuring semantic similarities applying ontology matching techniques.…”

Prediction of Datasets sameAs Interlinking on Web of Data

“…On Web of Data, the use of owl: sameAs [2] predicate is ubiquitous in interlinking Linked Data datasets to support merging distributed descriptions of equivalent RDF resources from different datasets. The considerable scale makes choosing target datasets that should be interlinked with a given dataset a big challenge [3,4]. In most related works [5,6], the authors didn't distinguish the RDF link types, while in real application scenarios, merely identifying target datasets for interlinking without specifying the link type would be of less help as dataset publishers still don't know what kinds of RDF links can be established and furthermore how to configure the data linking algorithms.…”

“…In most related works [5,6], the authors didn't distinguish the RDF link types, while in real application scenarios, merely identifying target datasets for interlinking without specifying the link type would be of less help as dataset publishers still don't know what kinds of RDF links can be established and furthermore how to configure the data linking algorithms. Nikolov et al [3] proposed approach of identifying relevant datasets for establish equivalence relations between individuals of datasets depending on the support of a third-party semantic search engine, and they have only conducted experiments on three datasets as examples.…”