This thesis analyzes the role of human error in cybersecurity, focusing on data breaches and incidents related to it. Using data from the VERIS Community Database, interarrival times were analyzed across six economic sectors, and the fit into exponential and geometric distributions was verified through goodness-of-fit tests. The results show that the exponential model, particularly in its two-parameter form, effectively describes the phenomenon when considering appropriate time windows. Analysis of the coefficient of variation squared also highlights the presence of bursty behavior. Finally, the thesis points out some limitations related to data coverage and completeness and proposes possible future applications for cyber risk management.
This thesis analyzes the role of human error in cybersecurity, focusing on data breaches and incidents related to it. Using data from the VERIS Community Database, interarrival times were analyzed across six economic sectors, and the fit into exponential and geometric distributions was verified through goodness-of-fit tests. The results show that the exponential model, particularly in its two-parameter form, effectively describes the phenomenon when considering appropriate time windows. Analysis of the coefficient of variation squared also highlights the presence of bursty behavior. Finally, the thesis points out some limitations related to data coverage and completeness and proposes possible future applications for cyber risk management.
STATISTICAL STUDY OF CYBERSECURITY INCIDENTS INTERARRIVAL TIMES: ANALYSIS OF DATA BREACHES CAUSED BY HUMAN ERROR
BUTTIGLIONE, PIETRO
2025/2026
Abstract
This thesis analyzes the role of human error in cybersecurity, focusing on data breaches and incidents related to it. Using data from the VERIS Community Database, interarrival times were analyzed across six economic sectors, and the fit into exponential and geometric distributions was verified through goodness-of-fit tests. The results show that the exponential model, particularly in its two-parameter form, effectively describes the phenomenon when considering appropriate time windows. Analysis of the coefficient of variation squared also highlights the presence of bursty behavior. Finally, the thesis points out some limitations related to data coverage and completeness and proposes possible future applications for cyber risk management.| File | Dimensione | Formato | |
|---|---|---|---|
|
pietrobuttiglione_tesi_1010.pdf
File consultabile da postazioni biblioteca
Descrizione: Relazione finale tesi
Dimensione
9.56 MB
Formato
Adobe PDF
|
9.56 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14253/110599