logo SBA

ETD

Archivio digitale delle tesi discusse presso l’Università di Pisa

Tesi etd-02052024-183745


Tipo di tesi
Tesi di laurea magistrale
Autore
DI VIRGILIO, DAVIDE
URN
etd-02052024-183745
Titolo
From Hikari-2021 Dataset to an Intrusion Detection Model
Dipartimento
INFORMATICA
Corso di studi
DATA SCIENCE AND BUSINESS INFORMATICS
Relatori
relatore Prof. Baiardi, Fabrizio
tutor Dott. Gatta, Marco
Parole chiave
  • classification
  • data mining
  • data science
  • intrusion detection model
  • machine learning
  • network traffic analysis
Data inizio appello
23/02/2024
Consultabilità
Non consultabile
Data di rilascio
23/02/2064
Riassunto

The document outlines the creation and analysis of a dataset named Hikari-2021, focusing on generating network data for cybersecurity purposes. After describing the methodology for dataset generation and various traffic profiles (background traffic, benign profile, and attacker profile), it delves into data preparation processes, including data understanding, cleaning, outlier detection, and feature selection. The issue of learning from imbalanced data is also addressed. The classification phase involves various techniques, including cross-validation and hyperparameter tuning for boosting algorithms such as Gradient Boosting, XGBoost, and bagging techniques like Random Forest. Finally, the last chapter discusses evaluation metrics and classifier performance.
File