A survey on data quality: principles, taxonomies and comparison of approaches.

Abstract

Nowadays, data generation keeps increasing exponentially due to the emergence of the Internet of Things (IoT) and Big data technologies. The manipulation of such Big amount of data becomes more and more difficult because of its size and its variety. For better governance of organizations (decision making, data analysis, earnings increase …), data quality and data governance at present of Big data are two major pillars for the design of any system handling data within the organization. This explains the number of researches conducted as it constitutes a research subject with several gaps and opportunities. Many works were conducted to define and standardize Data Quality (DQ) and its dimensions, others were directed to design and propose data quality assessment and improvement models or frameworks. This work aims to recall the data quality principles starting by the needed background knowledge, then identify and compare the relevant taxonomies existing in the literature, next surveys and compares the available Data quality assessment and improvement approaches. After that, we propose a metamodel highlighting the main concepts of DQ assessment and we describe a generic process for DQ assessment and improvement. Finally, we evoke the main challenges in the field of DQ before and after the emergence of Big Data.