Is There A "Golden" Rule for Code Reviewer Recommendation? : —An Experimental Evaluation

Hu, Yue; Wang, Junjie; Hou, Jie; Li, Shoubin; Wang, Qing

doi:10.1109/qrs51102.2020.00069

Cited by 7 publications

(3 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When small samples are taken as the historical review data, the first several iterations of the incremental sampling process may have relatively low performance and then decrease the final average top-k accuracy. This conjecture corroborates the recent findings by Hu et al [11] that the investigated code reviewer approaches are sensitive to training data on evaluation metrics.…”

Section: Neutronsupporting

confidence: 92%

“…In general, the results show that combining three similarity detection methods can relatively get the best performance of top-k accuracy and MRR on the four projects. Discussion of RQ1: The experiment results in Table 2 indicate that the selected similarity detection methods can produce acceptable performance scores (MRR values between 0.06 and 0.36) on code reviewer recommendation for architecture violations on the four projects, compared to the results (MRR values between 0.14 and 0.59) of related studies on generic reviewer recommendation (e.g., [7,11]) with more reviewer candidates (which means potentially better performance due to the larger datasets). Besides, we observed that the similarity detection methods can achieve varying performances on different OSS projects.…”

Section: Results and Discussion 41 Rq1: Effectiveness Of Our Approachmentioning

confidence: 95%

See 1 more Smart Citation

Code Reviewer Recommendation for Architecture Violations: An Exploratory Study

Peng

Avgeriou

2023

Proceedings of the 27th International Conference on Evaluation and Assessment in Software Engineering

View full text Add to dashboard Cite

Code review is a common practice in software development and often conducted before code changes are merged into the code repository. A number of approaches for automatically recommending appropriate reviewers have been proposed to match such code changes to pertinent reviewers. However, such approaches are generic, i.e., they do not focus on specific types of issues during code reviews. In this paper, we propose an approach that focuses on architecture violations, one of the most critical type of issues identified during code review. Specifically, we aim at automating the recommendation of code reviewers, who are potentially qualified to review architecture violations, based on reviews of code changes. To this end, we selected three common similarity detection methods to measure the file path similarity of code commits and the semantic similarity of review comments. We conducted a series of experiments on finding the appropriate reviewers through evaluating and comparing these similarity detection methods in separate and combined ways with the baseline reviewer recommendation approach, RevFinder. The results show that the common similarity detection methods can produce acceptable performance scores and achieve a better performance than RevFinder. The sampling techniques used in recommending code reviewers can impact the performance of reviewer recommendation approaches. We also discuss the potential implications of our findings for both researchers and practitioners. CCS CONCEPTS• Software and its engineering → Designing software; • General and reference → Empirical studies.

show abstract

Section: Neutronsupporting

confidence: 92%

Section: Results and Discussion 41 Rq1: Effectiveness Of Our Approachmentioning

confidence: 95%

Code Reviewer Recommendation for Architecture Violations: An Exploratory Study

Peng

Avgeriou

2023

Proceedings of the 27th International Conference on Evaluation and Assessment in Software Engineering

View full text Add to dashboard Cite

show abstract

“…When small samples are taken as the historical review data, the first several iterations of the incremental sampling process may have relatively low performance and then decrease the final average top-k accuracy. This conjecture corroborates the recent findings by Hu et al (Hu et al, 2020) that the investigated code reviewer approaches are sensitive to training data on evaluation metrics.…”

Section: Rq3: Comparison Of Sampling Methodssupporting

confidence: 91%

Understanding, Analysis, and Handling of Software Architecture Erosion

View full text Add to dashboard Cite

Architecture erosion reflects the tendency of an implemented architecture of a software system to gradually diverge from the intended architecture. Empirical evidence from numerous studies has demonstrated that architecture erosion significantly impacts various aspects of software development, maintenance, and evolution. This is because, architecture erosion often tends to occur imperceptibly and accumulates over time, making its repair challenging, costly, and sometimes even impossible. Therefore, timely detection and remediation of architecture erosion become crucial.One way to manage architecture erosion is by identifying its early symptoms, such as lack of modularity and various architectural smells. By identifying and managing these symptoms and their evolution, developers can gain insights into the software system's health and take proactive measures like architecture refactoring. This early warning mechanism not only aids in repairing eroded architecture, but also helps in understanding, identifying, analyzing, and optimizing software architecture, ultimately improving software product quality. However, despite several research studies investigating the architecture erosion phenomenon, the current state of the art has certain shortcomings. Specifically, there is a lack of comprehensive understanding of the nature of architecture erosion, it is unclear which symptoms of architecture erosion are the most common, and there is a lack of effective methods to identify these symptoms. Hence, the main research objective of this thesis is to establish a landscape of architecture erosion, investigate common erosion symptoms, and propose feasible approaches to identify and handle architecture erosion.To achieve the stated objective, we first need to obtain a landscape of the architecture erosion phenomenon and its current state of research in the literature. To this end, we conducted a systematic mapping study that covers the literature spanning te meten. Door middel van een reeks experimenten hebben we deze methoden afzonderlijk geëvalueerd en vergeleken met de basisbenadering (RevFinder). De resultaten toonden aan dat de gangbare gelijkenisdetectiemethoden acceptabele prestatiescores vertoonden en RevFinder overtroffen bij het aanbevelen van code reviewers voor architectuurschendingen. Bovendien ontdekten we dat de samplingtechnieken die worden gebruikt bij het aanbevelen van code reviewers, invloed hebben op de prestaties van reviewer aanbevelingsbenaderingen.

show abstract

Assessing Code Review Quality with ChatGPT: A Survey of Automated Reviewer Assignment Methods and Experimental Outcomes

Zydroń,

Protasiewicz

2024

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

Is There A "Golden" Rule for Code Reviewer Recommendation? : —An Experimental Evaluation

Cited by 7 publications

References 38 publications

Code Reviewer Recommendation for Architecture Violations: An Exploratory Study

Code Reviewer Recommendation for Architecture Violations: An Exploratory Study

Understanding, Analysis, and Handling of Software Architecture Erosion

Assessing Code Review Quality with ChatGPT: A Survey of Automated Reviewer Assignment Methods and Experimental Outcomes

Contact Info

Product

Resources

About