From big data to smart data: a data quality perspective.

Big Data (BD) solutions are designed to better support decision-making processes in order to optimize organizational performance. These BD solutions use company’s core business data, using typically large datasets. However, data that doesn’t meet adequate quality levels will lead to BD solutions that will not produce useful results, and consequently may not be used to make adequate business decisions. For a long time, companies have collected and stored large amounts of data without being able to exploit the advantage of exploring it. Nowadays, and thanks to the Big Data explosion, organizations have begun to recognize the need for estimating the value of their data and, vice-versa, managing data accordingly to their value. This need of managing the Value of data has led to the concept of Smart Data. It not only involves the datasets, but also the set of technologies, tools, processes and methodologies that enable all the Values from the data to the End-users (Business, data scientist, BI…). Consequently, Smart data is data actionable. We discovered that data quality is one of the most important issues when it comes to “smartizing” data. In this paper, we introduce a methodology to make data smarter, taking as a reference point, the quality level of the data itself.