Ontologies nowadays have become widely used for knowledge representation, and are considered as foundation Semantic Web.However with their wide spread usage, a question of evaluation increased even more.This paper addresses the issue finding an efficient ontology method by presenting existing techniques, while discussing advantages drawbacks.The presented techniques can be grouped into four categories: gold standard-based, corpus-based, task-based criteria based approaches.

Large knowledge graphs capture information of a large number entities and their relations. Among the many relations they capture, class subsumption assertions are usually present expressed using \texttt{rdfs:subClassOf} construct. From our examination, publicly available contain potentially erroneous cyclic subclass relations, problem that can be exacerbated when different integrated as Linked Open Data. In this paper, we an automatic approach for resolving such cycles at scale automated...

Most of the Linked Data applications currently rely on use owl: sameAs for linking ontology instances. However, several studies have noticed multiple misuses this identity link. These misuses, which are mainly caused by lack other well-defined alternatives, can lead to erroneous statements or inconsistencies. We propose in paper a new contextual link: identiConTo that could serve as replacement identical instances specified context. To detect these links, we defined an algorithm named DECIDE...

A century ago, the 1918–19 influenza pandemic swept across globe, taking lives of over 50 million people. We use data from Dutch civil registry to show which regions in Netherlands were most affected by pandemic. do so for entire 1918 year as well first, second, and third wave that hit summer 1918, autumn winter 1919. Our analyses excess mortality was highest Oost-Brabant, Zuid-Limburg, Noord-Holland, Drenthe, Groningen, Overijssel, whereas low Zuid-Beveland, Utrechtse Heuvelrug, Achterhoek....

In the absence of a central naming authority on Semantic Web, it is common for different data sets to refer same thing by names. Whenever multiple names are used denote thing, owl:sameAs statements needed in order link and foster reuse. Studies that date back as far 2009, observed property sometimes incorrectly. our previous work, we presented an identity graph containing over 500 million explicit 35 billion implied statements, scalable approach automatically calculating error degree each...

Data linking in knowledge graphs is a crucial and long-standing problem; it involves finding identity links between different descriptions that refer to the same real-world entity. Keys, which are subsets of properties identify each entity, important elements these links. The classical key-based approach data based on discovery set keys graph, then apply merge operation keys. But this can be very time consuming lead few valid two graphs. In work, we study issue transferability discovered one...

The goal of giving information a well-defined meaning is currently shared by different research communities. Once has meaning, it can be searched and retrieved more effectively. Therefore, this paper survey about the methods that compare textual sources in order to determine whether they address similar or not. improvement studied will eventually lead increase efficiency documentary research. In achieve goal, first category focuses on semantic measure definitions. A second paraphrase...

Different studies have observed that the semantic web identity predicate owl:SameAs is sometimes used incorrectly. In this paper, we show how network metrics such as community structure of graph can be in order to detect possibly erroneous statements. One benefit here presented approach it applied links, and does not rely on any additional knowledge. We evaluate our 558M statements scraped from LOD cloud. This evaluation shows ability scale, its efficiency detecting links.

In a decentralised knowledge representation system such as the Web of Data, it is common and indeed desirable for different graphs to overlap. Whenever multiple names are used denote same thing, owl:sameAs statements needed in order link data foster reuse. Whilst deductive value identity can be extremely useful enhancing various knowledge-based systems, incorrect use have wide-ranging effects global space like Data. With several works already proven that broken, this survey investigates...

LINKS stands for 'LINKing System historical family reconstruction' and is a software system to link nominal data from the Dutch archives ultimately reconstruct individuals families. We present background philosophy of this matching explain its structure functioning. Currently core consists indexed civil certificates. These certificates are available 1812 — start Vital Registration until year they confidential based on privacy laws. For more than 20 years, thousands volunteers have been...