Roland Croft

M. Mehdi Kholoosi, M. Ali Babar, Roland Croft

Artificial Intelligence (AI) advancements have enabled the development of Large Language Models (LLMs) that can perform a variety of tasks with remarkable semantic understanding and accuracy. ChatGPT is one such LLM that has gained significant attention due to its impressive capabilities for assisting in various knowledge-intensive tasks. Due to the knowledge-intensive nature of engineering secure software, ChatGPT's assistance is expected to be explored for security-related tasks during the development/evolution of software. To gain an understanding of the potential of ChatGPT as an emerging technology for supporting software security, we adopted a two-fold approach. Initially, we performed an empirical study to analyse the perceptions of those who had explored the use of ChatGPT for security tasks and shared their views on Twitter. It was determined that security practitioners view ChatGPT as beneficial for various software security tasks, including vulnerability detection, information retrieval, and penetration testing. Secondly, we designed an experiment aimed at investigating the practicality of this technology when deployed as an oracle in real-world settings. In particular, we focused on vulnerability detection and qualitatively examined ChatGPT outputs for given prompts within this prominent software security task. Based on our analysis, responses from ChatGPT in this task are largely filled with generic security information and may not be appropriate for industry use. To prevent data leakage, we performed this analysis on a vulnerability dataset compiled after the OpenAI data cut-off date from real-world projects covering 40 distinct vulnerability types and 12 programming languages. We assert that the findings from this study would contribute to future research aimed at developing and evaluating LLMs dedicated to software security.

Roland Croft, Brian Du, Darcy Joseph et al.

Face Recognition (FR) models have been shown to be vulnerable to adversarial examples that subtly alter benign facial images, exposing blind spots in these systems, as well as protecting user privacy. End-to-end FR systems first obtain preprocessed faces from diverse facial imagery prior to computing the similarity of the deep feature embeddings. Whilst face preprocessing is a critical component of FR systems, and hence adversarial attacks against them, we observe that this preprocessing is often overlooked in blackbox settings. Our study seeks to investigate the transferability of several out-of-the-box state-of-the-art adversarial attacks against FR when applied against different preprocessing techniques used in a blackbox setting. We observe that the choice of face detection model can degrade the attack success rate by up to 78%, whereas choice of interpolation method during downsampling has relatively minimal impacts. Furthermore, we find that the requirement for facial preprocessing even degrades attack strength in a whitebox setting, due to the unintended interaction of produced noise vectors against face detection models. Based on these findings, we propose a preprocessing-invariant method using input transformations that improves the transferability of the studied attacks by up to 27%. Our findings highlight the importance of preprocessing in FR systems, and the need for its consideration towards improving the adversarial generalisation of facial adversarial examples.

Roland Croft, M. Ali Babar, Li Li

Software Vulnerability (SV) severity assessment is a vital task for informing SV remediation and triage. Ranking of SV severity scores is often used to advise prioritization of patching efforts. However, severity assessment is a difficult and subjective manual task that relies on expertise, knowledge, and standardized reporting schemes. Consequently, different data sources that perform independent analysis may provide conflicting severity rankings. Inconsistency across these data sources affects the reliability of severity assessment data, and can consequently impact SV prioritization and fixing. In this study, we investigate severity ranking inconsistencies over the SV reporting lifecycle. Our analysis helps characterize the nature of this problem, identify correlated factors, and determine the impacts of inconsistency on downstream tasks. Our findings observe that SV severity often lacks consideration or is underestimated during initial reporting, and such SVs consequently receive lower prioritization. We identify six potential attributes that are correlated to this misjudgment, and show that inconsistency in severity reporting schemes can severely degrade the performance of downstream severity prediction by up to 77%. Our findings help raise awareness of SV severity data inconsistencies and draw attention to this data quality problem. These insights can help developers better consider SV severity data sources, and improve the reliability of consequent SV prioritization. Furthermore, we encourage researchers to provide more attention to SV severity data selection.

Roland Croft, Yongzheng Xie, M. Ali Babar

Software Vulnerability Prediction (SVP) is a data-driven technique for software quality assurance that has recently gained considerable attention in the Software Engineering research community. However, the difficulties of preparing Software Vulnerability (SV) related data is considered as the main barrier to industrial adoption of SVP approaches. Given the increasing, but dispersed, literature on this topic, it is needed and timely to systematically select, review, and synthesize the relevant peer-reviewed papers reporting the existing SV data preparation techniques and challenges. We have carried out a Systematic Literature Review (SLR) of SVP research in order to develop a systematized body of knowledge of the data preparation challenges, solutions, and the needed research. Our review of the 61 relevant papers has enabled us to develop a taxonomy of data preparation for SVP related challenges. We have analyzed the identified challenges and available solutions using the proposed taxonomy. Our analysis of the state of the art has enabled us identify the opportunities for future research. This review also provides a set of recommendations for researchers and practitioners of SVP approaches.

Triet H. M. Le, David Hin, Roland Croft et al.

It is increasingly suggested to identify Software Vulnerabilities (SVs) in code commits to give early warnings about potential security risks. However, there is a lack of effort to assess vulnerability-contributing commits right after they are detected to provide timely information about the exploitability, impact and severity of SVs. Such information is important to plan and prioritize the mitigation for the identified SVs. We propose a novel Deep multi-task learning model, DeepCVA, to automate seven Commit-level Vulnerability Assessment tasks simultaneously based on Common Vulnerability Scoring System (CVSS) metrics. We conduct large-scale experiments on 1,229 vulnerability-contributing commits containing 542 different SVs in 246 real-world software projects to evaluate the effectiveness and efficiency of our model. We show that DeepCVA is the best-performing model with 38% to 59.8% higher Matthews Correlation Coefficient than many supervised and unsupervised baseline models. DeepCVA also requires 6.3 times less training and validation time than seven cumulative assessment models, leading to significantly less model maintenance cost as well. Overall, DeepCVA presents the first effective and efficient solution to automatically assess SVs early in software systems.

Roland Croft, Yongzheng Xie, Mansooreh Zahedi et al.

Given programming languages can provide different types and levels of security support, it is critically important to consider security aspects while selecting programming languages for developing software systems. Inadequate consideration of security in the choice of a programming language may lead to potential ramifications for secure development. Whilst theoretical analysis of the supposed security properties of different programming languages has been conducted, there has been relatively little effort to empirically explore the actual security challenges experienced by developers. We have performed a large-scale study of the security challenges of 15 programming languages by quantitatively and qualitatively analysing the developers' discussions from Stack Overflow and GitHub. By leveraging topic modelling, we have derived a taxonomy of 18 major security challenges for 6 topic categories. We have also conducted comparative analysis to understand how the identified challenges vary regarding the different programming languages and data sources. Our findings suggest that the challenges and their characteristics differ substantially for different programming languages and data sources, i.e., Stack Overflow and GitHub. The findings provide evidence-based insights and understanding of security challenges related to different programming languages to software professionals (i.e., practitioners or researchers). The reported taxonomy of security challenges can assist both practitioners and researchers in better understanding and traversing the secure development landscape. This study highlights the importance of the choice of technology, e.g., programming language, in secure software engineering. Hence, the findings are expected to motivate practitioners to consider the potential impact of the choice of programming languages on software security.

Roland Croft, Dominic Newlands, Ziyu Chen et al.

Background: Static Application Security Testing (SAST) tools purport to assist developers in detecting security issues in source code. These tools typically use rule-based approaches to scan source code for security vulnerabilities. However, due to the significant shortcomings of these tools (i.e., high false positive rates), learning-based approaches for Software Vulnerability Prediction (SVP) are becoming a popular approach. Aims: Despite the similar objectives of these two approaches, their comparative value is unexplored. We provide an empirical analysis of SAST tools and SVP models, to identify their relative capabilities for source code security analysis. Method: We evaluate the detection and assessment performance of several common SAST tools and SVP models on a variety of vulnerability datasets. We further assess the viability and potential benefits of combining the two approaches. Results: SAST tools and SVP models provide similar detection capabilities, but SVP models exhibit better overall performance for both detection and assessment. Unification of the two approaches is difficult due to lacking synergies. Conclusions: Our study generates 12 main findings which provide insights into the capabilities and synergy of these two approaches. Through these observations we provide recommendations for use and improvement.

Triet H. M. Le, Roland Croft, David Hin et al.

Context: Security Vulnerabilities (SVs) pose many serious threats to software systems. Developers usually seek solutions to addressing these SVs on developer Question and Answer (Q&A) websites. However, there is still little known about on-going SV-specific discussions on different developer Q&A sites. Objective: We present a large-scale empirical study to understand developers' SV discussions and how these discussions are being supported by Q&A sites. Method: We first curate 71,329 SV posts from two large Q&A sites, namely Stack Overflow (SO) and Security StackExchange (SSE). We then use topic modeling to uncover the topics of SV-related discussions and analyze the popularity, difficulty, and level of expertise for each topic. We also perform a qualitative analysis to identify the types of solutions to SV-related questions. Results: We identify 13 main SV discussion topics on Q&A sites. Many topics do not follow the distributions and trends in expert-based security sources such as Common Weakness Enumeration (CWE) and Open Web Application Security Project (OWASP). We also discover that SV discussions attract more experts to answer than many other domains, but some difficult SV topics (e.g., Vulnerability Scanning Tools) still receive quite limited support from experts. Moreover, we identify seven key types of answers given to SV questions on Q&A sites, in which SO often provides code and instructions, while SSE usually gives experience-based advice and explanations. Conclusion: Our findings provide support for researchers and practitioners to effectively acquire, share and leverage SV knowledge on Q&A sites.

Triet H. M. Le, David Hin, Roland Croft et al.

Security is an increasing concern in software development. Developer Question and Answer (Q&A) websites provide a large amount of security discussion. Existing studies have used human-defined rules to mine security discussions, but these works still miss many posts, which may lead to an incomplete analysis of the security practices reported on Q&A websites. Traditional supervised Machine Learning methods can automate the mining process; however, the required negative (non-security) class is too expensive to obtain. We propose a novel learning framework, PUMiner, to automatically mine security posts from Q&A websites. PUMiner builds a context-aware embedding model to extract features of the posts, and then develops a two-stage PU model to identify security content using the labelled Positive and Unlabelled posts. We evaluate PUMiner on more than 17.2 million posts on Stack Overflow and 52,611 posts on Security StackExchange. We show that PUMiner is effective with the validation performance of at least 0.85 across all model configurations. Moreover, Matthews Correlation Coefficient (MCC) of PUMiner is 0.906, 0.534 and 0.084 points higher than one-class SVM, positive-similarity filtering, and one-stage PU models on unseen testing posts, respectively. PUMiner also performs well with an MCC of 0.745 for scenarios where string matching totally fails. Even when the ratio of the labelled positive posts to the unlabelled ones is only 1:100, PUMiner still achieves a strong MCC of 0.65, which is 160% better than fully-supervised learning. Using PUMiner, we provide the largest and up-to-date security content on Q&A websites for practitioners and researchers.