Hands-On Artificial Intelligence for Cybersecurity

User detection with multilayer perceptrons

Why does the MLP classifier show considerably better results in terms of prediction accuracy?

The answer lies in the fact that it represents an artificial neural network (ANN).

ANNs constitute the fundamental element of deep learning and are at the base of high potential that characterizes deep learning algorithms, allowing, for example, the classification of enormous amounts of data, the performance of face and speech recognition, or beating a world chess champion such as Kasparov.

We met the perceptron in Chapter 3, Ham or Spam? Detecting Email Cybersecurity Threats with AI; we have also seen its limitations regarding the classification scenarios in which data isn't linearly separable. However, ...