datagrade - Data Quality Assessment on the ISO/IEC 25012 Model
A single-call data quality assessment scoring a data set
on the five inherent characteristics of the ISO/IEC 25012 data
quality model (accuracy, completeness, consistency, credibility
and currentness) through the fifteen quality properties beneath
them, ISO/IEC (2008) <https://www.iso.org/standard/35736.html>.
Each property is measured as a ratio of counted data items, the
form ISO/IEC 25024 measurement functions take, ISO/IEC (2015)
<https://www.iso.org/standard/35749.html>. Six properties are
computable from a table alone; the rest compare the data
against expectations declared through a specification object,
and are reported as not applicable rather than approximated
when none is supplied. Also reports missing-data patterns,
duplicates, redundant columns, outliers, multicollinearity and
distributional deviations, and returns a structured object with
plots rather than console text alone. The statistics implement
the methodology of the author's master's dissertation, Demirhan
(2024) "Assessing Data Quality Management Issues in
Open-Sourced Data Sets with Statistical Methodology"
<https://halukdemirhan.com/assets/msc-thesis-haluk-demirhan-public.pdf>.