Various software vulnerabilities classifications have been constructed since the early 70s for correct understanding of vulnerabilities, and thus acts as a strong foundation to protect and prevent software from exploi...
详细信息
ISBN:
(纸本)9783642221903
Various software vulnerabilities classifications have been constructed since the early 70s for correct understanding of vulnerabilities, and thus acts as a strong foundation to protect and prevent software from exploitation. However, despite all research efforts, exploitable vulnerabilities still exist in most major software, the most common still being C overflows vulnerabilities. C overflow vulnerabilities are the most frequent vulnerabilities to appear in various advisories with high impact or critical severity. Partially but significantly, this is due to the absence of a sourcecode perspective taxonomy to address all types of C overflow vulnerabilities. Therefore, we propose this taxonomy, which also classifies the latest C overflow vulnerabilities into four new categories. We also describe ways to detect and overcome these vulnerabilities. and hence, acts as a valuable reference for developers and security analysts to identify potential security C loopholes so as to reduce or prevent exploitations altogether.
Detecting source code vulnerabilities is an essential issue today. In this paper, to improve the efficiency of detecting vulnerabilities in software written in C/C++, we propose to use a combination of Deep Graph Conv...
详细信息
Detecting source code vulnerabilities is an essential issue today. In this paper, to improve the efficiency of detecting vulnerabilities in software written in C/C++, we propose to use a combination of Deep Graph Convolutional Neural Network (DGCNN) and code property graph (CPG). Specifically, 3 main proposed phases in the research method include: phase 1: building feature profiles of sourcecode. At this step, we suggest using analysis techniques such as Word2vec, one hot encoding to standardize and analyze the sourcecode;phase 2: extracting features of sourcecode based on feature profiles. Accordingly, at this phase, we propose to use Deep Graph Convolutional Neural Network (DGCNN) model to analyze and extract features of the sourcecode;phase 3: classifying sourcecode based on the features extracted in phase 2 to find normal sourcecode and sourcecode containing security vulnerabilities. Some scenarios for comparing and evaluating the proposed method in this study compared with other approaches we have taken show the superior effectiveness of our approach. Besides, this result proves that our method in this paper is not only correct and reasonable, but it also opens up a new approach to the task of detecting source code vulnerabilities.
暂无评论