Paper Type
Complete
Abstract
The paper presents a vulnerability detection system based on deep learning that automatically learns features from a large set of real-world code. The dataset was collected from Github and contains Python code with various vulnerabilities, which has been adapted for CuBERT model. Data samples were created from the source code of vulnerable files by analyzing individual code tokens and their context, allowing for a detailed analysis. Pre-trained CuBERT model is fine-tuned on samples for each type of vulnerability to recognize the characteristics of vulnerable code and is then applied to detect vulnerabilities in source code.
Paper Number
2089
Recommended Citation
Poniszewska-Marańda, Aneta; Krasnowski, Wojciech; and Cegielski, Marcin, "Recognizing security patterns in code using machine learning – Python case study" (2025). AMCIS 2025 Proceedings. 1.
https://aisel.aisnet.org/amcis2025/sig_odis/sig_odis/1
Recognizing security patterns in code using machine learning – Python case study
The paper presents a vulnerability detection system based on deep learning that automatically learns features from a large set of real-world code. The dataset was collected from Github and contains Python code with various vulnerabilities, which has been adapted for CuBERT model. Data samples were created from the source code of vulnerable files by analyzing individual code tokens and their context, allowing for a detailed analysis. Pre-trained CuBERT model is fine-tuned on samples for each type of vulnerability to recognize the characteristics of vulnerable code and is then applied to detect vulnerabilities in source code.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
SIGODIS