Dendritic cell algorithm applied to ping scan investigation revisited: detection quality and performance analysis
Carregando...
Data
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal de Minas Gerais
Descrição
Tipo
Artigo de periódico
Título alternativo
Primeiro orientador
Membros da banca
Resumo
The dendritic cell algorithm is an immune inspired method based on the danger model, which relies on cell interactions to antigens and signals, considering the correlation between both events, to solve anomaly detection problems. Starting with new datasets, comprising ping scans and file transfers in computer networks, this paper proposes improvements in the algorithm test methodology and qualitative analysis. Results measurement, detection capability, and performance evaluation are explored and discussed, including a real-time analysis and a comparison with similar approaches. Concluding, the study discusses advantages and limitations of the studied approach, suggesting possible improvements and new applications.
Abstract
Assunto
Aprendizado do computador, Inteligência artificial
Palavras-chave
Algorithm design and analysis , Immune system , Signal processing algorithms , Intrusion detection , Computational intelligence , Ports (Computers), Artificial immune systems , dendritic cell algorithm , intrusion detection , ping scan investigation, Digital Networks , Detection Capability , Anomaly Detection , False Positive , False Negative , Support Vector Machine , False Positive Rate , Normal Process , Input Signal , Adaptive Algorithm , Normal Operation , False Negative Rate , Apoptotic Signaling , Intrusion Detection , Knowledge Extraction , Real-time Simulation , Immune Responses In Models , Attack Scenarios , Intrusion Detection System , Classical Version , Deterministic Version , Normal Scenario , Mixed Scenario , Cohen’s Kappa , Scanning Process , Computational Intelligence , Performance Metrics , Suppression Of Immune Responses
Citação
Departamento
Curso
Endereço externo
https://ieeexplore.ieee.org/document/8006369