Toward a reliability measurement framework automated using deep learning

Toward a reliability measurement framework automated using deep learning
复制标题

使用深度学习实现自动化的可靠性测量框架

DOI:
10.1145/3314058.3317733
复制
发表时间:
2019
期刊:
Proceedings of the 6th Annual Symposium on Hot Topics in the Science of Security
影响因子:
--
通讯作者:
Niu, Jianwei
Niu, Jianwei
中科院分区:
--
文献类型:
--
作者:
Heaps, John;Zhang, Xueling;Wang, Xiaoyin;Breaux, Travis;Niu, Jianwei

文献摘要

相似文献

提出了一种基于代码模式检测的软件缺陷检测框架。我们的框架将挖掘和生成错误模式,检测代码中的这些模式,并计算软件的脆弱性度量。虽然我们的框架表现良好,但我们意识到它需要繁重的手动任务,这使得该框架在实践中无法使用。然而,我们相信机器学习的最新进展将使我们能够将深度学习技术应用于源代码,这将有助于自动化我们的框架,以便在真实的世界中实现更好的实用性。
We propose a framework to detect software bugs based on code pattern detection. Our framework will mine and generate bug patterns, detect those patterns in code, and calculate a vulnerability measure of software. While our framework performs well, we realize that it requires heavy manual tasks that render the framework infeasible to use in practice. However, we believe that recent advancements in machine learning will allow us to apply deep learning techniques to source code, which will help automate our framework for better practicality in the real world.