Assessing Data Quality of Integrated Data by Quality Aggregation of its Ancestors

Abstract

Data Quality is degraded during the process of extracting and merging data from multiple heterogeneous sources. Besides, users have no information regarding the quality of the accessed data.

This document presents the methods utilized to assess data quality at multiple levels of granularity, including derived non-atomic data, considering data provenance.  The Data Quality Manager prototype has been implemented and tested  to prove such assessment.