Metodyczne podejście do analizy i eksploracji danych marketingowych

Jerzy Korczak , Maciej Pondel

Abstract

The article proposes a methodology for development of a marketing Decision Support System using data mining methods and Big Data technologies. The main research findings focus on the analysis and exploration of very large, heterogeneous sets of highly volatile marketing data. The approach is inspired by the CRISP-DM methodology which is not oriented towards Big Data applications. The article describes in detail the stages of the project development according to the extended CRISP-DM methodology, taking into account the specificity of the analysis and exploration processes of large marketing databases processed in real time. In order to illustrate the approach, the examples based on real data about customers, transactions and products of the Internet store were discussed