Vulnerability Time Series Forecasting: A Comparative Study of Hierarchical and Non-Hierarchical Approaches

Cybersecurity has become an increasing priority in large organizations due to the rapid evolution of digital threats. With increasingly sophisticated attacks, vulnerability management emerges as one of the main strategies to protect systems and data. However, the complexity and dynamics of threats m...

Full description

Saved in:
Bibliographic Details
Main Authors: Juscimara G. Avelino, Bruna A. O. Brito, Thiago T. de Oliveira, Jean M. M. de Lima, Andre M. Gurgel, Ramon S. Malaquias, Itamir de M. B. Filho
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10945367/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849702138599440384
author Juscimara G. Avelino
Bruna A. O. Brito
Thiago T. de Oliveira
Jean M. M. de Lima
Andre M. Gurgel
Ramon S. Malaquias
Itamir de M. B. Filho
author_facet Juscimara G. Avelino
Bruna A. O. Brito
Thiago T. de Oliveira
Jean M. M. de Lima
Andre M. Gurgel
Ramon S. Malaquias
Itamir de M. B. Filho
author_sort Juscimara G. Avelino
collection DOAJ
description Cybersecurity has become an increasing priority in large organizations due to the rapid evolution of digital threats. With increasingly sophisticated attacks, vulnerability management emerges as one of the main strategies to protect systems and data. However, the complexity and dynamics of threats make management processes challenging, requiring efficient and adaptable approaches. Artificial Intelligence (AI) and Machine Learning (ML) are promising tools to anticipate and mitigate these threats, offering predictive capabilities based on historical data. This article proposes the use of time series in a hierarchical manner for the forecasting of vulnerabilities in systems. The proposed methodology aims to deal with the complexity of the data, allowing a hierarchical structure to understand and capture the interdependencies and patterns among different levels more completely. The evaluation is carried out with different ML models, such as LSTM, RNN, MLP, among others, comparing the performance of hierarchical and non-hierarchical approaches. The results indicate that, in a hierarchical structure, especially LSTM, shows superior performance in vulnerability forecasting. However, in some scenarios, when using reconcilers, models like N-BEATS also demonstrate good results.
format Article
id doaj-art-dd9ed0cbd4a74cb7a5acac7d48ffd27f
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-dd9ed0cbd4a74cb7a5acac7d48ffd27f2025-08-20T03:17:44ZengIEEEIEEE Access2169-35362025-01-0113586585867010.1109/ACCESS.2025.355586010945367Vulnerability Time Series Forecasting: A Comparative Study of Hierarchical and Non-Hierarchical ApproachesJuscimara G. Avelino0https://orcid.org/0000-0002-6934-0746Bruna A. O. Brito1https://orcid.org/0009-0001-8116-495XThiago T. de Oliveira2https://orcid.org/0009-0006-7457-5941Jean M. M. de Lima3https://orcid.org/0000-0002-2324-9365Andre M. Gurgel4https://orcid.org/0000-0002-1925-8031Ramon S. Malaquias5https://orcid.org/0000-0002-8350-2836Itamir de M. B. Filho6https://orcid.org/0000-0003-1694-8237Digital Metropolis Institute, Federal University of Rio Grande do Norte, Natal, Rio Grande do Norte, BrazilDigital Metropolis Institute, Federal University of Rio Grande do Norte, Natal, Rio Grande do Norte, BrazilDigital Metropolis Institute, Federal University of Rio Grande do Norte, Natal, Rio Grande do Norte, BrazilDigital Metropolis Institute, Federal University of Rio Grande do Norte, Natal, Rio Grande do Norte, BrazilDigital Metropolis Institute, Federal University of Rio Grande do Norte, Natal, Rio Grande do Norte, BrazilDigital Metropolis Institute, Federal University of Rio Grande do Norte, Natal, Rio Grande do Norte, BrazilDigital Metropolis Institute, Federal University of Rio Grande do Norte, Natal, Rio Grande do Norte, BrazilCybersecurity has become an increasing priority in large organizations due to the rapid evolution of digital threats. With increasingly sophisticated attacks, vulnerability management emerges as one of the main strategies to protect systems and data. However, the complexity and dynamics of threats make management processes challenging, requiring efficient and adaptable approaches. Artificial Intelligence (AI) and Machine Learning (ML) are promising tools to anticipate and mitigate these threats, offering predictive capabilities based on historical data. This article proposes the use of time series in a hierarchical manner for the forecasting of vulnerabilities in systems. The proposed methodology aims to deal with the complexity of the data, allowing a hierarchical structure to understand and capture the interdependencies and patterns among different levels more completely. The evaluation is carried out with different ML models, such as LSTM, RNN, MLP, among others, comparing the performance of hierarchical and non-hierarchical approaches. The results indicate that, in a hierarchical structure, especially LSTM, shows superior performance in vulnerability forecasting. However, in some scenarios, when using reconcilers, models like N-BEATS also demonstrate good results.https://ieeexplore.ieee.org/document/10945367/Cybersecuritymachine learninghierarchical time seriesvulnerability forecasting
spellingShingle Juscimara G. Avelino
Bruna A. O. Brito
Thiago T. de Oliveira
Jean M. M. de Lima
Andre M. Gurgel
Ramon S. Malaquias
Itamir de M. B. Filho
Vulnerability Time Series Forecasting: A Comparative Study of Hierarchical and Non-Hierarchical Approaches
IEEE Access
Cybersecurity
machine learning
hierarchical time series
vulnerability forecasting
title Vulnerability Time Series Forecasting: A Comparative Study of Hierarchical and Non-Hierarchical Approaches
title_full Vulnerability Time Series Forecasting: A Comparative Study of Hierarchical and Non-Hierarchical Approaches
title_fullStr Vulnerability Time Series Forecasting: A Comparative Study of Hierarchical and Non-Hierarchical Approaches
title_full_unstemmed Vulnerability Time Series Forecasting: A Comparative Study of Hierarchical and Non-Hierarchical Approaches
title_short Vulnerability Time Series Forecasting: A Comparative Study of Hierarchical and Non-Hierarchical Approaches
title_sort vulnerability time series forecasting a comparative study of hierarchical and non hierarchical approaches
topic Cybersecurity
machine learning
hierarchical time series
vulnerability forecasting
url https://ieeexplore.ieee.org/document/10945367/
work_keys_str_mv AT juscimaragavelino vulnerabilitytimeseriesforecastingacomparativestudyofhierarchicalandnonhierarchicalapproaches
AT brunaaobrito vulnerabilitytimeseriesforecastingacomparativestudyofhierarchicalandnonhierarchicalapproaches
AT thiagotdeoliveira vulnerabilitytimeseriesforecastingacomparativestudyofhierarchicalandnonhierarchicalapproaches
AT jeanmmdelima vulnerabilitytimeseriesforecastingacomparativestudyofhierarchicalandnonhierarchicalapproaches
AT andremgurgel vulnerabilitytimeseriesforecastingacomparativestudyofhierarchicalandnonhierarchicalapproaches
AT ramonsmalaquias vulnerabilitytimeseriesforecastingacomparativestudyofhierarchicalandnonhierarchicalapproaches
AT itamirdembfilho vulnerabilitytimeseriesforecastingacomparativestudyofhierarchicalandnonhierarchicalapproaches