Preparação de dados de plano de saúde suplementar para algoritmos de mineração de dados

Abstract

The large accumulation of data generated by information systems that are essential in all areas makes possible the process of Knowledge Discovery in Database (KDD), but these data need to be adjusted in some way it is possible to extract some knowledge potentially useful for the domain in which it applies. Thus, to obtain knowledge of large databases is used the metologias of KDD. In this work we used the methodology proposed by Fayyad-1996b with this methodology was carried out pre-processing part of the data warehouse of a Prepaid Health Plans in order to include records contained in a particular table as an attribute of a new table, grouping the events that have been made by beneficiaries of Prepaid Health Plans. At the end applying all of the proposed Methodology and going through all the steps, it was reached as a result the creation of a tool to manipulate the entire process making it possible to convert the events of records attributes of a table.