Retrieving and ranking scientific publications from linked open data repositories

Abstract

Standard-Nutzungsbedingungen: Dieses Dokument darf zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen dieses Dokument nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, aufführen, vertreiben oder anderweitig nutzen. Sofern für das Dokument eine Open-Content-Lizenz verwendet wurde, so gelten abweichend von diesen Nutzungsbedingungen die in der Lizenz gewährten Nutzungsrechte.

Semantic Web Science and Real-World Applications

“…• Path-based algorithms use information about semantic paths within a graph structure to compute similarities useful to produce recommendations. For example, spreading activation (Cheekula, Kapanipathi, Doran, & Jain, 2015;Chicaiza, Piedra, López-Vargas, & Tovar-Edmundo, 2014;Hajra, Latif, & Tochtermann, 2014;Marie, Gandon, Legrand, & Ribière, 2013), random walk (Cantador, Konstas, & Jose, 2011); path-weights for vertex discovery (Strobin & Niewiadomski, 2014). Modern methods combine machine learning to learn the best path to consider relying on learning to rank (Di Noia, Ostuni, Tomeo, & Di Sciascio, 2016)…”

Allied

“…• Path-based: use information about semantic paths within a RDF graph structure to compute distances. Spreading activation (Marie, Gandon, Legrand, & Ribi ere, 2013;Hajra et al, 2014;Cheekula, Kapanipathi, Doran, & Jain, 2015); random walk (Cantador, Konstas, & Jose, 2011); and path-weights for vertex discovery (Strobin & Niewiadomski, 2014). Table 1, shows a comparison between graph-based algorithms.…”

Enriching Scientific Publications from LOD Repositories Through Word Embeddings Approach

“…Based on our previous evaluation conducted using 112 publications, the list of retrieved publications according to the aligned concepts between repositories was extremely wide [8]. For example, in order to deliver more details, the concept "biofuel" from EconStor is aligned to Agrovoc as "biofuels", and is used for describing 7083 documents in OpenAgris catalog.…”

“…Therefore, we also use alignments between repositories/thesauruses for retrieving an initial set of publications, especially for reformulating a search query from one vocabulary to another [8]. The presence of thesauri in the primary and targeting repository can be useful for extending the corpus of metadata concepts, which, as we will show later, is very significant for further analyses.…”