Automated Severity Driven Patch Management

  • Carlos Eduardo de Schuller Banjar UFRJ
  • Miguel Angelo Santos Bicudo UFRJ
  • Lucas Miranda UFRJ
  • Cainã Figueiredo Pereira UFRJ
  • Lucas Senos Coutinho UFRJ
  • Daniel Sadoc Menasche UFRJ
  • Gaurav Kumar Srivastava Siemens
  • Enrico Lovat Siemens
  • Anton Kocheturov Siemens
  • Matheus Martins Siemens
  • Leandro Pfleger de Aguiar Siemens

Abstract


We present a method for assessing the temporal severity associated with software vulnerabilities by analyzing reported vulnerability data. Data from various platforms is collected and curated to define specific vulnerability features and historical vulnerability event data. When a vulnerability is specified, the system identifies its vulnerability class using a classifier based on the predefined features. Historical event data is then processed to generate a predictive severity curve, which estimates the evolution of a temporal severity score, parameterized by the occurrence of key vulnerability events. This curve predicts the time of weaponization and/or exploitation events, along with the corresponding severity score for the specified vulnerability. Our approach aims to support automated decision-making in software patch management by enabling accurate tracking and prediction of vulnerability severity over time.
Keywords: Severity assessment, vulnerabilities, exploits, CVSS
Published
2024-11-26
How to Cite

Select a Format
BANJAR, Carlos Eduardo de Schuller et al. Automated Severity Driven Patch Management. In: INDUSTRY TRACK - LATIN-AMERICAN SYMPOSIUM ON DEPENDABLE COMPUTING (LADC), 13. , 2024, Recife/PE. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2024 . p. 179–183.