Semantic of Data Dependencies to Improve the Data Quality.

Data quality in databases is a critical challenge because the cost of anomalies may be very high, especially for large databases. Therefore, the correction of these anomalies represents an issue that has become more and more important both in enterprises and in academia. In this work, we address the problems of intra-column and inter-columns anomalies in big data. We propose a new approach for data cleaning that takes into account the semantic dependencies between the columns of a data source. The novelty of our proposal is the reduction of the size of the search space in the process of functional dependency discovery based on data semantics. In this paper, we present the first steps of our work. They allow recognizing the semantics of data and correct intra-column anomalies.