Context-Aware Software Vulnerability Classification Using Machine Learning

Siewruk, Grzegorz; Mazurczyk, Wojciech

doi:10.1109/access.2021.3075385

Cited by 12 publications

(8 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When comparing the experiment results between Table 5 and Table 3, our proposed method was more effective than other approaches. Specifically, Russell's approach [8] was inferior to ours on all measurements. Especially with the recall score, this method was lower than our proposed method by more than 60%.…”

Section: Experiments Results Of Scenariomentioning

confidence: 59%

“…In this paper, we reuse the experiment results presented in the study [34]. Specifically, the approaches compared and evaluated with our method include: REVEAL [34], Russell [8], VulDeePecker [11], SySeVR [31], and Devign [46]. Table 5 below lists the experiment results of other approaches on the Verum dataset.…”

Section: Experiments Results Of Scenariomentioning

confidence: 99%

“…Moreover, the DrCSE was also much more optimized when experimented on FFMpeg+Qemu Dataset. For example, when comparing DrCSE with Russell [8] , SySeVR [31] , Devign [45] , the DrCSE model was between 6% and 25% more efficient. Likewise, VulDeePecker [11] was less effective than DrCSE from 10% to 35% on all measurements.…”

Section: Experiments Results Of Scenariomentioning

confidence: 99%

“…Recently, most approaches often focus on analyzing source code in the semantic and syntactic form and then using machine learning or deep learning algorithms to analyze and predict vulnerabilities. However, we believe that the current approaches are facing some of the following problems [3,4,5,6,7,8,9,10,11,53]:…”

Section: Problemmentioning

confidence: 99%

See 3 more Smart Citations

DrCSE: A New Framework For Software Vulnerability Detection base on An Advanced Computing

Cong,

Xuan

2023

Preprint

View full text Add to dashboard Cite

The detection of software vulnerabilities written in C and C++ languages takes a lot of attention and interest today. This paper proposes a new framework called DrCSE to improve software vulnerability detection. It uses an intelligent computation tech-nique based on the combination of two methods: rebalancing data and representation learning to analyze and evaluate the code property graph (CPG) of the source code for detecting abnormal behavior of software vulnerabilities. To do that, DrCSE performs a combination of 3 main processing techniques: i) building the source code feature profiles, ii) rebalancing data, and iii) contrastive learning. In which, the method i) extracts the source code’s features based on the vertices and edges of the CPG. The method of rebalancing data has the function of supporting the training process by balancing the experimental dataset. Finally, contrastive learning techniques learn the important features of the source code by finding and pulling similar ones together while pushing the outliers away. The experiment part of this paper demonstrates the superiority of the DrCSE Framework for detecting source code security vulnerabil-ities using the Verum dataset. As a result, the method proposed in the article has brought a pretty good performance in all metrics, especially the Precision and Recall scores of 39.35% and 69.07%, respectively, proving the eﬃciency of the DrCSE Framework. It performs better than other approaches, with a 5% boost in Precision and a 5% boost in Recall. Overall, this is considered the best research result for the software vulnerability detection problem using the Verum dataset according to our survey to date

show abstract

Section: Experiments Results Of Scenariomentioning

confidence: 59%

Section: Experiments Results Of Scenariomentioning

confidence: 99%

Section: Experiments Results Of Scenariomentioning

confidence: 99%

Section: Problemmentioning

confidence: 99%

See 2 more Smart Citations

DrCSE: A New Framework For Software Vulnerability Detection base on An Advanced Computing

Cong,

Xuan

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…As presented above, for the problem of source code vulnerability detection, the two most critical factors determining the accuracy of the classification process are feature extraction and classification processes. However, based on our observations, recent approaches have not effectively addressed both these issues, leading to limited experimental results [8,9]. Specifically, as follows:…”

Section: Motivationmentioning

confidence: 94%

A New Framework for Software Vulnerability Detection Based on an Advanced Computing

Cong,

Xuan

2024

CMC

View full text Add to dashboard Cite

The detection of software vulnerabilities written in C and C++ languages takes a lot of attention and interest today. This paper proposes a new framework called DrCSE to improve software vulnerability detection. It uses an intelligent computation technique based on the combination of two methods: Rebalancing data and representation learning to analyze and evaluate the code property graph (CPG) of the source code for detecting abnormal behavior of software vulnerabilities. To do that, DrCSE performs a combination of 3 main processing techniques: (i) building the source code feature profiles, (ii) rebalancing data, and (iii) contrastive learning. In which, the method (i) extracts the source code's features based on the vertices and edges of the CPG. The method of rebalancing data has the function of supporting the training process by balancing the experimental dataset. Finally, contrastive learning techniques learn the important features of the source code by finding and pulling similar ones together while pushing the outliers away. The experiment part of this paper demonstrates the superiority of the DrCSE Framework for detecting source code security vulnerabilities using the Verum dataset. As a result, the method proposed in the article has brought a pretty good performance in all metrics, especially the Precision and Recall scores of 39.35% and 69.07%, respectively, proving the efficiency of the DrCSE Framework. It performs better than other approaches, with a 5% boost in Precision and a 5% boost in Recall. Overall, this is considered the best research result for the software vulnerability detection problem using the Verum dataset according to our survey to date.

show abstract

Tool-Based Approach on Digital Vulnerability Management Hub (VMH) by Using TheHive Platform

Sharmila,

Sitharthaa,

Babu

et al. 2023

Springer Proceedings in Mathematics &Amp; Statistics

View full text Add to dashboard Cite

Context-Aware Software Vulnerability Classification Using Machine Learning

Cited by 12 publications

References 30 publications

DrCSE: A New Framework For Software Vulnerability Detection base on An Advanced Computing

DrCSE: A New Framework For Software Vulnerability Detection base on An Advanced Computing

A New Framework for Software Vulnerability Detection Based on an Advanced Computing

Tool-Based Approach on Digital Vulnerability Management Hub (VMH) by Using TheHive Platform

Contact Info

Product

Resources

About