A Machine Learning Based Approach for Detection of Distributed Denial of Service Attacks

Venkata, Raghavender Kotla

doi:10.1007/978-981-19-4831-2_7

Cited by 1 publication

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Kotla Venkata (2022) [7] explored the application of machine learning in detecting Distributed Denial of Service (DDoS) attacks. The author proposed a machine learning-based approach aimed at efficiently identifying such attacks, which are characterized by overwhelming a network or service with excessive requests.…”

Section: Literature Reviewmentioning

confidence: 99%

“…In synthesizing the literature, it is evident that the research landscape is characterized by a diverse array of machine learning techniques, each tailored to address specific challenges in network intrusion detection. While some studies focus on specific attack vectors, such as DDoS attacks (Kotla Venkata, 2022) [7] or SQL Injection vulnerabilities (Zhang, 2019) [15], others propose comprehensive solutions that encompass various attack types (Vinayakumar et al, 2022 [9]; Gurina & Eliseev, 2019 [12]). The comparative analysis also reveals a gradual shift from traditional machine learning techniques to more sophisticated deep learning and ensemble approaches (Wang et al, 2020 [10]; Ravi et al, 2022 [11]).…”

Section: Comparative Analysismentioning

confidence: 99%

See 1 more Smart Citation

A Machine Learning-based Approach for Detecting Network Intrusions in Large-scale Networks

Venkata Krishna,

Venkatesh Sharma,

MallaReddy

2023

IJCERT

View full text Add to dashboard Cite

The objective of this research is to explore and enhance the mechanisms for detecting network intrusions, particularly focusing on large-scale networks. Traditional Intrusion Detection Systems (IDS) are frequently challenged by several limitations. These include high rates of false alarms, an inability to adapt swiftly to new and evolving threats, and challenges in scaling to accommodate large volumes of network traffic. Addressing these limitations, the study introduces a comprehensive approach that incorporates machine learning techniques to bolster network security. The methodology specifically employs Support Vector Machines (SVM) and Decision Trees as classifiers. SVM is known for its effectiveness in classifying high-dimensional data, while Decision Trees are favoured for their ease of interpretation and decision-making transparency. The research meticulously evaluates and contrasts the proposed approach with existing IDS. It reveals that the integration of SVM and Decision Trees significantly improves the accuracy of intrusion detection, with the model achieving an accuracy rate of up to 95% in certain test scenarios. This marks a substantial enhancement compared to traditional IDS. Furthermore, the study emphasizes the model's capability to adapt in real-time to emerging threats. This adaptability ensures that the IDS remains robust and effective even as network threats evolve, thereby addressing a critical gap in existing systems. In conclusion, this research underscores the potential of machine learning, specifically through the use of SVM and Decision Trees, in enhancing the precision, adaptability, and scalability of intrusion detection systems in large-scale networks. The findings suggest that such an approach can mitigate prevalent challenges in network security and contribute significantly to establishing a more secure and resilient cyber environment.

show abstract

Section: Literature Reviewmentioning

confidence: 99%