Assessing the Quality of owl: sameAs Links.

owl:sameAs is one of the most important properties of the Linked Open Data domain. The property is used to indicate that two things are the same and when these two things come from two different datasets. But unfortunately, there is a gap between the purpose of this property, the way it was conceived and designed, and the way it is used in the wild. This deficiency is mainly due to the misuse of owl: sameAs because its strict semantics is not always taken into account or well understood and there may be no clear alternative. As a result, inaccurate data can be either crawled or inferred from incorrect (or at least questionable) links. Depending on the need for high quality data, the impact on the end users of the data can be considerable. This is why we propose to study how erroneous links can influence data, how to repair such cases, or even better, how to prevent the production of these links. We want to assess wrong links impact (e.g. at which points they decrease overall quality) and how they can be addressed.