Query approximation in the case of incompletely aligned datasets

Clouds of Linked Open Data about the same domain promote the formulation of a query over a source dataset and then try to process the same query over different target datasets, one after another, in order to obtain a broader set of answers. However, heterogeneity of vocabularies used in the datasets and the scarce number of alignments among those datasets makes that querying task extremely difficult. This paper presents a proposal that allows on demand transformations of queries by using a set of transformation rules that are able to rewrite a query formulated over a source dataset into another query adequate for a target dataset, which approximates the original one. The approach relieves users from knowing the vocabulary used in the targeted datasets and even more it considers situations where alignments do not exist or they are not suitable for the formulated query. Therefore, in order to favor the possibility of getting answers, sometimes there is no guarantee of obtaining a semantically equivalent translation. Experiments with benchmark queries validate the feasibility of the proposal.