This work was the code employed for my thesis for the Master of Science in Data Science and Management at Luiss Guido Carli University.
The primary objectives of this project are as follows:
To construct an Intrusion Detection System (IDS) through the execution of classification tasks using a range of machine learning and deep learning techniques. These tasks involve distinguishing between binary representations of attacks and normal activities, as well as performing multiclass classification to identify the specific type of attack.
Subsequently, these same algorithms were applied to a modified dataset, simulating two distinct types of data poisoning with varying levels of intensity. The purpose of this exercise is to analyze the effects of these modifications on the performance of the models.
By pursuing these goals, the project aims to enhance our understanding of IDS systems and their resilience to adversarial attacks, ultimately contributing to the improvement of cybersecurity measures.