Analysis of UNSW-NB15 Datasets Using Machine Learning Algorithms

Azeroual, Hakim; Belghiti, Imane Daha; Berbiche, Naoual

doi:10.1007/978-3-031-02447-4_21

Cited by 2 publications

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dataset contains traffic data from real network environments including 9 types of attacks such as Fuzzers, Analysi, Backdoors, Exploits, Generic, etc. Azeroual, H. et al cleaned the data and identified the most important features, evaluated multiple machine learning algorithms using PCA and yielded effective results for attack classification with the proposed DT-PCA model [3] . The CIC-IDS series dataset was created by the Cybersecurity Lab at Carleton University, Canada.…”

Section: Introductionmentioning

confidence: 99%

Real-time traffic intrusion detection based on CNN-LSTM deep neural networks

Liu,

Ma,

Gao

et al. 2024

International Conference on Computer Network Security and Software Engineering (CNSSE 2024)

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Real-time traffic intrusion detection based on CNN-LSTM deep neural networks

Liu,

Ma,

Gao

et al. 2024

International Conference on Computer Network Security and Software Engineering (CNSSE 2024)

View full text Add to dashboard Cite

Contribution to Threat Management Through the Use of AI-Based IDS

Saadi,

Belghiti,

Atbib

et al. 2024

RGSA

View full text Add to dashboard Cite

Objectives: This paper aims to enhance cybersecurity through the integration of Artificial Intelligence (AI) in Intrusion Detection Systems (IDS), addressing the limitations of traditional IDS in detecting evolving cyber threats. Theoretical Framework: The study builds on existing research in cybersecurity, focusing on AI techniques such as decision trees and linear regression to improve the accuracy and effectiveness of AI-based IDS. Method: A comprehensive review of current AI-IDS methodologies is conducted, alongside an exploration of machine learning algorithms applied to datasets like KDD99 and NSL-KDD. The proposed architecture utilizes supervised machine learning to predict anomalies in network traffic. Results and Discussion: The findings indicate that AI-IDS can significantly reduce false positives and enhance detection of zero-day attacks through adaptive learning. The results highlight the importance of quality data and continuous model refinement. Research Implications: This research underscores the necessity for ongoing exploration of AI techniques in cybersecurity, suggesting future studies focus on real-time adaptive systems to further improve threat detection. Originality/Value: This paper contributes to the field by providing insights into the practical application of AI in IDS, offering a structured approach that combines theoretical knowledge with empirical evidence, thus paving the way for future innovations in cybersecurity.

show abstract

Analysis of UNSW-NB15 Datasets Using Machine Learning Algorithms

Cited by 2 publications

References 9 publications

Real-time traffic intrusion detection based on CNN-LSTM deep neural networks

Real-time traffic intrusion detection based on CNN-LSTM deep neural networks

Contribution to Threat Management Through the Use of AI-Based IDS

Contact Info

Product

Resources

About