The Investigative Study on the Performance Analysis of SMOTE Employed Machine Learning Classifier Models to DDoS Attack Detection - Analytical Research

Aug 5, 2026

Distributed Denial of Service (DDoS) attacks are active network threats that disrupt online services and cause major losses in finance and reputation. This study uses machine learning models enhanced with Synthetic Minority Over-sampling Technique (SMOTE) to address imbalanced DDoS datasets, specifically CSE-CIC-2018. Five algorithms—Naive Bayes, Random Forest, Logistic Regression, Decision Tree, and XGBoost—are evaluated for detection performance. Results show Random Forest achieving the strongest balance of F1-Score (0.99), MCC (0.98), and accuracy (0.99), providing guidance for selecting models while managing accuracy versus computational efficiency.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages7

Reading Time18 min

AccessFree to download

Introduction

Background on DDoS attacks

DDoS attack types and classification

DDoS detection approaches and defenses

Proposed Study Overview

Dataset and SMOTE-based handling

Machine learning classifiers evaluated

Results and Comparative Performance

Evaluation metrics and findings

Best-performing model insights

Conclusion

Practical guidance and tradeoffs

The Investigative Study on the Performance Analysis of SMOTE employed Machine Learning Classifier Models to DDoS Attack Detection [1] Sravan Kumar G, [2] Dr. M Sunitha, [3] Ghantasala Srinivasa Rithik, [4] S Veeresh Kumar, [5] Dr K Sreerama Murthy [1] CVR College of Engineering, Hyderabad, [2] CVR College of Engineering, Hyderabad, [3] CVR College of Engineering, Hyderabad, [4] St. Martin’s Engineering College, Secunderabad, [5] Koneru Lakshmaiah Education Foundation, Hyderabad. [1][sravankumarcvr@gmail.com](sravankumarcvr@gmail.com) ,[2] [palemonisunitha@gmail.com](palemonisunitha@gmail.com), [3] [srithik2002@gmail.com](srithik2002@gmail.com), [4][salvadiveeresh2023@gmail.com](salvadiveeresh2023@gmail.com), [5][drsreeram1203@gmail.com](drsreeram1203@gmail.com) Abstract—Distributed Denial of Service (DDoS) attack, a severe attack on the network services during the contemporary era, is categorized under active attacks in security attacks. The impact of this attack on the organization or individual resources leads to massive loss in terms of finance, reputation. Therefore, detecting Distributed DDoS attacks is vital in ensuring the availability and integrity of online services of an organization. The work in this paper employed machine learning techniques, complemented by Synthetic Minority Over-sampling Technique (SMOTE), to tackle the inherent challenge of imbalanced DDoS attack dataset: CSE-CIC-2018 and to enhance computational efficiency while maintaining accuracy with a fraction of the original dataset. The emphasis of the this works is to comprehensively assess the performance of five prominent algorithms of machine learning-Naive Bayes, Random Forest, Logistic Regression, Decision Tree, and XGBoost-in the context of detection of DDoS attack. The overhead of oversampling is handled with the application of SMOTE oversampling and it has been addressed data imbalance issues, improving the algorithms' capability to identify attacks of DDoS effectively. The work of this paper finds and reveals distinct comparative advantages among the algorithms employed in the DDoS attack detection and provides actionable insights in choosing the most suitable algorithms of Machine learning for the detection of DDoS attack, provided emphasizing the significance of SMOTE to enhance the algorithms' performance in the presence of imbalanced data. Eventually, this paper offers invaluable guidance for organizations seeking to make safe their network against DDoS attacks while considering the crucial tradeoffs between accuracy and computational efficiency. The proposed work in this paper presented the results that Random Forest classifier ensured the better performance with F1-Score value 0.99, Mathews Correlation Coefficient (MCC) value 0.98 and accuracy value 0.99 relative to other classifiers employed. Keywords- Distributed Denial of Service, DDoS attack, Machine Learning, SMOTE, Naïve Bayes, Random Forest, Logistic Regression, Decision Tree, XGBoost, Mathews Correlation Coefficient, F1-Score. I. INTRODUCTION In terms of resources and time, an active attack has vital implications for IT infrastructure. Cyber attacks that include attack of DDoS causes financial losses to organizations and businesses. The most hazardous attack is DDoS attacks and it has been produced literature in this area [11, 12, 13] . These attacks directly affect the economic and financial sector. For example, the Mirai attack in October, was a series of DDOS attacks targeting the "DNS"; supplier Dyn and its operations on October 21, 2016 [11, 12, 13] . Attacks of these kind resulted service interruption of platforms of the Internet over multiple regions across North American and European nations [11, 12, 13] . The first DDoS attack that cut off all Internet access in a city for several hours occurred in 1997 at the hacker conference in Las Vegas by attacker Khan C Smith [14] . After this attack, many online attacks took place against Sprint, EarthLink, ETrad

FAQs

Why is detecting DDoS attacks important for organizations?

DDoS attacks disrupt network services and can lead to substantial financial losses and reputational damage, making detection essential to preserve availability and integrity of online systems.

How does this study address the imbalanced DDoS dataset problem?

It applies SMOTE oversampling to mitigate class imbalance in the CSE-CIC-2018 dataset, improving the classifiers’ ability to recognize DDoS attacks.

Which classifiers are evaluated for DDoS attack detection?

The paper evaluates five machine learning algorithms: Naive Bayes, Random Forest, Logistic Regression, Decision Tree, and XGBoost.

What performance did the best model achieve in the study?

Random Forest delivered the strongest reported performance, with F1-Score 0.99, MCC 0.98, and accuracy 0.99 compared with the other classifiers.

Aug 5, 2026

Upload by Emma Wilson