Bayesian Network (BN)–based Learning Path Recommendation (LPR) systems are widely adopted in personalized education for modeling uncertainty and providing interpretable learner representations. However, existing studies predominantly evaluate these systems under controlled settings that assume balanced data, simplified curricula, and unconstrained resources. Consequently, limited empirical understanding exists regarding their performance in authentic classrooms. This study addresses this gap by examining the real-world deployment of a BN-based LPR system using naturally occurring classroom data. The system is evaluated using 426,004 quiz responses collected across 19 formative assessments,... More >

Skin‑disease classification has evolved from simple image recognizers into software‑driven pipelines that demand reliability, reproducibility, and ethical governance. While most AI reviews focus on algorithmic accuracy, few examine these systems through a software‑engineering (SE) lens—essential for assessing pipeline modularity, version control, deployment readiness, and long‑term maintainability, all critical for clinical integration. This review surveys literature from 2015 to early 2025, curating about 180 papers that link skin‑disease classification with SE practices. It traces the shift from handcrafted feature‑based classifiers to end‑to‑end convolutional, ensemble,... More >

Agile has become a cornerstone of modern software development. Among its many frameworks, Extreme Programming (XP) and Scrum are the most widely recognized. XP emphasizes technical practices and engineering discipline while Scrum provides structured roles and iterative planning. Over time, many organizations have also adopted hybrid models that combine the strengths of both. Despite their popularity, teams often face challenges when deciding which approach to adopt. The choice between XP, Scrum or a hybrid is not always straightforward as each carries different strengths, limitations and suitability for specific contexts. This paper addresses this issue by presenting a comparative analysis o... More >

Hypertension, a life-threatening global health challenge, requires early detection to prevent severe cardiovascular complications. Fundus imaging reveals microvascular alterations, yet conventional diagnosis often misses subtle early changes. This study introduces a multimodal deep learning framework that integrates clinical data, fundus images, and demographic features to improve hypertension prediction. Unlike single-modality approaches, our method captures complementary risk factors from both structured and unstructured data. We evaluate machine learning and deep learning models on clinical data, confirming DL's superior accuracy. For fundus images alone, a CNN achieves 74.44% accuracy, h... More >

The rise of AI tools such as GitHub Copilot and ChatGPT has reshaped software development by providing substantial support for coding and debugging tasks. Although these tools enhance productivity and reduce routine workload, existing research has largely emphasized short-term efficiency gains, leaving their long-term cognitive and pedagogical effects insufficiently explored. This study investigates the cognitive trade-offs associated with sustained reliance on generative AI, with particular attention to students and junior developers. Recent empirical findings indicate that excessive dependence on AI assistance may weaken deep debugging skills, impede conceptual understanding, and challenge... More >