Modelos estadísticos para la detección de patrones en medio ambiente y finanzas.

Abstract

In this work, some aspects of recurrent neural networks are implemented and studied in the field of detecting fraudulent transactions with credit cards. To do this, it begins with a review of neural networks. Then, the basic concepts of recurrent neural networks are developed and to illustrate it, an introduction to the problem of detecting fraud in credit card transactions is presented. Finally, a model of recurrent neural networks in Python is shown, in which the data treatment procedures are presented; such as the use of Keras’ library for the creation of a model and the results of the problem posed.