Prior Data Quality Management in Data Mining Process

Data Mining (DM) projects are implemented by following the knowledge discovery process. Several techniques for detecting and handling data quality problems such as missing data, outliers, inconsistent data or time-variant data, can be found in the literature of DM and Data Warehousing (DW). Tasks that are related to the quality of data are mostly in the Data Understanding and in the Data Preparation phases of the DM process. The main limitation in the application of the data quality management techniques is the complexity caused by a lack of anticipation in the detection and resolution of the problems. A DM process model designed for the prior management of data quality is proposed in this work. In this model, the DM process is defined in relation to the Software Engineering (SE) process; the two processes are combined in parallel. The main contribution of this DM process is the anticipation and the automation of all activities necessary to remove data quality problems.