Título: Defesa Inteligente: Detectando Ataques em Rede de Computadores com Aprendizado de Máquina
Autores: Wellington Resende de Araújo Júnior, Gabriel Vinicios Moreira Fernandes & Agnaldo José da Rocha Reis
Resumo: One proposes in this study an approach to detecting intrusions in computer networks through machine learning techniques, applying classification models, including decision trees, random forests and artificial neural networks, to the CICIDS2017 real dataset. The main objective is to identify and classify different types of cyber attacks, such as DoS, Web Attacks and infiltrations, by analyzing network traffic via Machine Learning models. The methodology employed comprises data pre-processing steps, selection of relevant attributes and model training using specific libraries, such as TensorFlow. The results obtained indicate that the DT model achieved superior performance levels, demonstrating high accuracy and recall close to 1, while the remaining models did not perform as good as the DT-based models.
Palavras-chave: Cybersecurity; Artificial Neural Networks; Intrusion Detection; Machine Learning; Classification Models.
Páginas: 6
Código DOI: 10.21528/CBIC2025-1191455
Artigo em PDF: CBIC_2025_paper1191455.pdf
Arquivo BibTeX:
CBIC_2025_1191455.bib