Resumen

ESPINOSA-ZUNIGA, Javier Jesús. Implementation of the CRISP-DM methodology for geographical segmentation using a public database. Ing. invest. y tecnol. [online]. 2020, vol.21, n.1, e00008.  Epub 03-Ago-2020. ISSN 2594-0732.  https://doi.org/10.22201/fi.25940732e.2020.21n1.008.

Technological progress has allowed to the organizations to store big amounts of data. However, organizations are facing to the challenge of analyzing such data for getting useful knowledge for decision making in real situations. Nowadays there are several methodologies that allow organizations to analyze big amounts of data in order to get information and knowledge. One of them is CRISP-DM (Cross Industry Standard Process for Data Mining) that despite the fact of be the most widely used methodology for Data Mining projects and to have more than twenty years old, it is yet not well known for many organizations in Mexico. This article aims to illustrate how to apply CRISP-DM for getting a geographical segmentation model for a public database called DENUE which contains a directory of business units in Mexico. The six steps of the methodology (understanding problem, understanding data, preparation of data, modeling, evaluation and implementation) has been applied in order to get a geographical segmentation model that divides Mexican geographical entities according to their business units. Albit some observations were classified not properly (according to the evaluation that was applied to the model) in general the clusters are acceptable considering the variables used for getting them, and in order to improve the model we suggest to consider additional variables that are no disposable in DENUE database nowadays. Although it is a segmentation model over DENUE database which is susceptible of improvement, it shows the potential of applying CRISP-DM for Data Mining projects and also shows the potential of exploiting public databases in order to get knowledge useful for many purposes (business, scholars, etc.).

Palabras llave : Segmentation; methodology; DENUE; CRISP-DM; datamining.

        · resumen en Español     · texto en Español     · Español ( pdf )