|
Acadêmico(a): Lucas Dalcol Pereira |
Título: MINERAÇÃO DE DADOS PARA GERAÇÃO DE ÁRVORE DE DECISÃO: APLICAÇÃO VENDAS DE VAREJO |
|
Abstract: |
The present work describes the process of modeling and developing a data mining application prototype on a substrate of a retail sales database. The work describes the steps of tabulating, transforming, cleaning and generating the data in an ARFF file for the application of the J48 algorithm, through the Java library provided by the Weka tool. The applied transformation process is presented and the final version of the decision tree is presented. The obtained results demonstrate the feasibility of the applied process and confirm the need for data processing and mining for the classification and determination of data standards to obtain information relevant to decision making, especially for gaining competitiveness in the retail sales market. However, for a more effective result, it is necessary to extend the time horizon of the data as well as to extend the scope involving, for example, data on the purchases made and the data on clients |
|
|