What does Dataset Reuse tell us about Quality?
Abstract
Following the Linked Data principles means maximising the
reusability of data over the Web. A strong reason for reusing a dataset is
that it is considered useful for some application. Considering the broad
definition of data quality as \fitness for use", the question arises whether
quality of linked datasets and their actual reuse correlate, or, in other
words, whether certain quality characteristics can be optimised to increase
the potential reuse of the datasets. Reuse of datasets becomes
apparent when datasets are referred to from other datasets, papers, or
discussions within the community. It can thus be measured, similarly to
citations of papers. Many other aspects of Linked Data quality have also
been defined in a measurable way, i.e. as quality metrics. In this paper
we present metrics to quantify dataset reuse in a scientific community
and investigate their correlation with the quality metrics discussed in the
literature.