Security system based on S.T.R.I.D.E using artificial intelligence

Abstract

In the last years hackers have strengthened and have taken advantage of their knowledge and skills, as well as the inappropriate use of technology to search for vulnerabilities in organizations, in order to obtain their own benefit or for third parties. It’s imperative to stop this type of attacks, looking for an alternative that allows to know them and learn to classify them, to be prepared for these contingencies. This project seeks to implement a neural network as protection in a computational network. This work will use a multilayer network feedforward that is the type of classifier based on neural networks most widely studied and used. A wide dataset is used for training it and the mean square error (MSE) for the error value measurement. Recently it has been established that neural networks are a promising alternative to several classification methods.