Code Authorship Attribution (CAA) has several applications such as copyright disputes, plagiarism detection and criminal prosecution. Existing studies mainly focused on CAA by proposing machine learning (ML) and Deep Learning (DL) based techniques. The main limitations of ML-based techniques are (a) manual feature engineering is required to train these models and (b) they are vulnerable to adversarial attack. In this study, we initially fine-tune five Large Language Models (LLMs) for CAA and evaluate their performance. Our results show that LLMs are robust and less vulnerable compared to existing techniques in CAA task.

Bug report management has been shown to be an important and time consuming software maintenance task. Often, the first step in managing bug reports is related to triaging a bug to the appropriate developer who is best suited to understand, localize, and fix the target bug. Additionally, assigning a given bug to a particular part of a software project can help to expedite the fixing process. However, despite the importance of these activities, they are quite challenging, where days can be spent on the manual triaging process. Past studies have attempted to leverage the limited textual data of bug reports to train text classification models that automate this process - to varying degrees of success. However, the textual representations and machine learning models used in prior work are limited by their expressiveness, often failing to capture nuanced textual patterns that might otherwise aid in the triaging process. Recently, large, transformer-based, pre-tained neural text representation techniques (i.e., large language models or LLMs) such as BERT and CodeBERT have achieved greater performance with simplified training procedures in several natural language processing tasks, including text classification. However, the potential for using these techniques to improve upon prior approaches for automated bug triaging is not well studied or understood. Therefore, in this paper we offer one of the first investigations that fine-tunes transformer-based language models for the task of bug triaging on four open source datasets, spanning a collective 53 years of development history with over 400 developers and over 150 software project components. Our study includes both a quantitative and qualitative analysis of effectiveness. Our findings illustrate that DeBERTa is the most effective technique across the triaging tasks of developer and component assignment, and the measured performance delta is statistically significant compared to other techniques. However, through our qualitative analysis, we also observe that each technique possesses unique abilities best suited to certain types of bug reports.

Authorship attribution identifies the true author of an unknown document. Authorship attribution plays a crucial role in plagiarism detection and blackmailer identification, however, the existing studies on authorship attribution in Bengali are limited. In this paper, we propose an instance-based deep authorship attribution model, called DAAB, to identify authors in Bengali. Our DAAB model fuses features from convolutional neural networks and another set of features from an artificial neural network to learn the stylometry of an author for authorship attribution. Extensive experiments with three real benchmark datasets such as Bengali-Quora and two online Bengali Corpus demonstrate the superiority of our authorship attribution model.

Move Method Refactoring (MMR) is used to place highly coupled methods in appropriate classes for making source code more cohesive. Like other refactoring techniques, it is mandatory that applying MMR will preserve applications’ behaviors. However, traditional MMR techniques failed to meet this essential precondition for Action methods in web-based application and API methods in libraries projects. The reason is that applying MMR on these methods changes the behaviors of the projects by raising Application-breaking issues, for instance, failure of browser requests and compilation errors in client projects. To resolve this problem, developers are suggested to manually check Action and API methods while applying MMR. However, manually inspecting thousands of lines of code for these issues is a time-consuming and hectic task. In this paper, an advanced MMR technique is proposed which automatically identifies Application-breaking MMR suggestions. This technique first takes the initial move method suggestions from the existing prominent MMR techniques e.g. JDeodorant. For each of the suggestions, it parses the source code and construct Abstract Syntax Tree to examine two types of usage. One is whether a suggestion has not been used in any unit test and Regular Class, and another is whether the suggestion has been used in unit test classes only. If any MMR suggestion is found having one of these two types of usage or both, the respective suggestion is marked as Application-breaking. In order to evaluate the proposed technique, several experiments have been conducted on open source projects. The experimental results show that the proposed technique achieved 96.4% Precision, 90% Recall and 93.1% F-score in detecting Application-breaking MMR suggestions, because of considering external dependencies of the MMR suggestions.