Journal of Software-Evolution and Process最新文献

Cross-project defect prediction (CPDP) poses a nontrivial challenge to construct a reliable defect predictor by leveraging data from other projects, particularly when data owners are concerned about data privacy. In recent years, federated learning (FL) has become an emerging paradigm to guarantee privacy information by collaborative training a global model among multiple parties without sharing raw data. While the direct application of FL to the CPDP task offers a promising solution to address privacy concerns, the data heterogeneity arising from proprietary projects across different companies or organizations will bring troubles for model training. In this paper, we study the privacy-preserving CPDP with data heterogeneity under the FL framework. To address this problem, we propose a novel knowledge enhancement approach named FedDP with two simple but effective solutions: 1. local heterogeneity awareness and 2. global knowledge distillation. Specifically, we employ open-source project data as the distillation dataset and optimize the global model with the heterogeneity-aware local model ensemble via knowledge distillation. Experimental results on 19 projects from two datasets demonstrate that our method significantly outperforms baselines.

Software quality models have increasingly emphasized human factors and user needs. In 2011, ISO/IEC 25010 introduced the quality-in-use (QinU) model, designed to evaluate software quality as an outcome of a user utilizing a system through the evaluation of five characteristics: effectiveness, efficiency, freedom from risk, satisfaction, and context coverage. As a generic reference, this standard has been customized for various software types (e.g., web portals and artificial intelligence systems). This article presents a customization for context-aware software systems (CASSs), which are software systems that interpret and use context information (regarding the user, the software application features, and the environment) to adapt their functionalities. We are particularly interested in CASS for pervasive, or ubiquitous, environments. To address this goal, each QinU characteristic was analyzed by professionals from the academy and industry, taking into account the CASS features for pervasive/ubiquitous environments. A cyclical process of definition, revision, and improvement based on measurement theory was carried out before empirical validation in case studies. As the main result, a novel set of QinU measures specifically tailored for CASSs in a pervasive environment is provided, considering not only the classic explicit user interactions (e.g., mouse clicks and text input) but also the implicit interactions during everyday activities (e.g., walking or driving), captured through sensors and processed to support the user (e.g., recommending nearby museums and providing driving guidance). This set of measures supports CASS assessment and improvements, offering more accurate and context-sensitive quality measurement.

Developing an information technology (IT) infrastructure for business and organizational operations has gained importance concurrently with the emergence of edge and cloud computing technologies. On the other hand, innovation is crucial for businesses because consumers demand better service and lower ownership costs. Several firms use the cloud as their innovation platform to expand and stay competitive. In the commercial world, cloud computing is swiftly embraced by small, medium, and big businesses in sectors including aerospace, automotive, logistics, financial services, textiles, and health. With all the benefits cloud computing has for business improvement, some issues and problems prevent organizations and companies from migrating to the cloud. Due to these evidentiary gaps, I combined quantitative and qualitative analytical techniques to synthesize the existing literature and identify possible directions for future studies that may have an impact. The primary aim of this research is to provide a bibliometric procedure to review the state of the art on this scope and provide a roadmap for future studies; then, it provides insight into how the adoption and sustained usage of cloud sourcing could and edge encourage the creation of novel business models and impact a company's competitive advantage. The publications were reviewed using a bibliometric approach that divided the papers into four categories: cloud computing data security, business risk management techniques, resource allocation, and business performance. According to the results, adopting cloud and edge computing service delivery models is connected with increased revenue and less capital investment in IT assets for the company. Also, the results showed that one of the main obstacles to implementing cloud and edge computing technology in organizations is low security and lack of sufficient internal resources.

Assessing the quality and credibility of research is crucial across disciplines. However, training early career scholars in systematic quality appraisal poses challenges. The rise of online grey literature increases the need for nuanced evaluation capabilities. This study aims to impart basic literature quality assessment knowledge in early career software engineering researchers using an interactive card-based learning activity. The PRISMA abstract quality checklist was adapted into a physical card deck. Sixteen novice researchers participated in a session using the cards to collaborate, discuss, and analyze a sample review paper abstract based on structured criteria. Quantitative feedback was gathered. Survey results indicated the card activity positively enhanced perceived understanding of quality principles, engagement, and assessment skills. Open-ended feedback highlighted cards improved focus, interactivity, and peer exchanges. This preliminary study provides encouraging evidence that a customized card-based approach can effectively instill foundational skills for assessing abstract quality while increasing motivation and enjoyment. Further research should evaluate long-term retention and optimal instructional design parameters.

Recent studies have shown that many software developers do not have sufficient knowledge and understanding of how to develop a privacy-friendly system. This may become a challenge in developing systems complying with data protection laws. To address this issue, we investigated the factors that influence developers' decision-making when developing privacy-sensitive systems. We conducted an empirical study by means of a survey with 109 practitioners. Our data analysis is based on the principles of social cognitive theory, which includes personal, behavioral, and external environmental factors. We identified six personal, five behavioral, and five external environment factors that affect how developers make decisions regarding privacy, including confusion between privacy and security and reliance on informal practices and organizational support gaps. These findings contribute to understanding how practitioners and companies consider privacy, showing improvements in formal training and structured support over previous studies yet highlighting persistent challenges in consistent privacy integration.

Software development effort estimation (SDEE) is a key task in managing software projects. Among the existing SDEE models, artificial neural networks (ANN) have garnered considerable attention from the software engineering community because of their ability to learn from previous data and yield acceptable estimates. However, to the best of the authors' knowledge, no systematic literature review (SLR) has been carried out with focus on the use of ANNs in SDEE. This work aims to analyze ANN-based SDEE studies from five view-points: estimation accuracy, accuracy comparison, estimation context, impact of combining ANN-based SDEE models with other techniques, and ANNs parameters. To find relevant ANN-based SDEE studies, we carried out an automated search using four electronic databases. The quality of the relevant papers was assessed to determine the set of papers to include in our review. We identified 65 papers published in the period 1993–2023 with acceptable quality score. The results of our systematic review revealed that ANN-based SDEE models perform better than 11 machine learning (ML) and non-ML SDEE models. Further, the estimation accuracy is improved when neural networks are used in combination with other techniques such as fuzzy clustering techniques. This study found that the use of ANN models in SDEE is promising to get accurate estimates. However, the application of ANN models in industry is still limited. Therefore, it is recommended that practitioners cooperate with researchers to encourage and facilitate the application of ANN models in industry.