- Research Article
- 10.1016/j.sasc.2026.200472
Optimizing machine learning models for obesity risk prediction through hyperparameter tuning
- Jun 01, 2026
- Systems and Soft Computing
- Salliah Shafi + 2 more +2
Publications from 2021 to 2026
Showing 10 of 186 papers
Optimizing machine learning models for obesity risk prediction through hyperparameter tuning
Discipline in Disguise: A Foucauldian Perspective on Agile at Scale
This study critically examines the transformation of Agile methodologies when scaled within large organizations, focusing on frameworks such as SAFe® and LeSS. Originally intended to foster flexibility, autonomy, and collaboration, Agile in its scaled form often reintroduces hierarchical controls, creating what is termed a paradox of “discipline in disguise.” While these frameworks claim to enhance empowerment and adaptability, they frequently institutionalize mechanisms of surveillance and normalization through rituals, metrics, and digital tools. Drawing on Michel Foucault’s concepts of disciplinary power, panopticism, and governmentality, this research analyzes how practices such as daily stand-ups, Agile boards, and coaching roles serve as instruments of subtle control. Findings suggest that transparency mechanisms—such as dashboards and progress metrics—promote self-monitoring and conformity, reducing psychological safety and innovation. Agile coaches further reinforce organizational norms under the guise of facilitation, creating decentralized yet pervasive governance. This paper contributes to critical management studies by reframing scaled Agile as both enabling and constraining. It argues for reflexive implementations that maintain Agile’s core values while mitigating disciplinary tendencies embedded in its structures. The analysis highlights the tension between autonomy and control, urging organizations to preserve adaptability without reinforcing bureaucratic practices.
Read moreEffect of Limestone Calcined Clay Cement (LC3) on Properties of Self-Compacting Concrete
تُعد الخرسانة ذاتية الدمك (Self-Compacting Concrete – SCC) من التقنيات المتقدمة في مجال الخرسانة، لما توفره من سهولة في التنفيذ وتحسين في المتانة، إلا أن استخدامها على نطاق واسع يظل محدودًا بسبب المحتوى العالي من الأسمنت واعتمادها على مواد إسمنتية مضافة مستوردة، لا سيما في ليبيا. في هذا السياق، برز إسمنت الحجر الجيري والطين المحروق (Limestone Calcined Clay Cement – LC3) كبديل واعد في المناطق التي تتوفر فيها خامات الطين الكلسي والحجر الجيري محليًا. تهدف هذه الدراسة إلى تقييم إمكانية استخدام LC3 المُنتج من مواد محلية في السياق الليبي ضمن الخرسانة ذاتية الدمك، وذلك كبديل جزئي للإسمنت البورتلاندي بنسبة 15% و30% و45%. أظهرت النتائج التجريبية أن الخرسانة ذاتية الدمك المعتمدة على LC3 تحقق متطلبات القبول الخاصة بالخواص الريولوجية، رغم الانخفاض التدريجي في قابلية الانسياب مع زيادة نسبة الاستبدال. كما بينت الاختبارات الميكانيكية أن الاستبدال المعتدل بـ LC3 يعزز المقاومة في الأعمار المبكرة، في حين تحقق نسب الاستبدال الأعلى مقاومة مكافئة أو محسّنة في الأعمار المتأخرة. ولم يُلاحظ أي تأثير سلبي على خصائص الشد حتى عند أعلى نسبة استبدال مقارنة بالخلطة المرجعية. إضافةً إلى ذلك، فإن انخفاض المسامية الكلية وازدياد سرعة نبضة الموجات فوق الصوتية يدلّان على تحسّن البنية المجهرية للخرسانة، مما ينعكس إيجابًا على أدائها الميكانيكي. وبوجه عام، تؤكد النتائج الجدوى التقنية لاستخدام LC3 في الخرسانة ذاتية الدمك حتى نسبة استبدال 45% بمواد محلية، مع تحقيق أداء مماثل أو أفضل من الخلطة الخرسانة ذاتية الدمك المرجعية.
Read moreRevolutionizing cancer treatment with senotherapeutics: a current perspective.
Cellular senescence is a double-edged sword in cancer biology, initially acting as a tumor-suppressive mechanism but later contributing to cancer progression and therapy resistance. Senescent cells, characterized by stable cell cycle arrest, secrete a complex array of bioactive molecules known as the senescence-associated secretory phenotype (SASP), which fosters chronic inflammation, disrupts tissue architecture, and promotes tumorigenesis through paracrine signaling. Accumulation of these cells in the tumor microenvironment can enhance malignancy, drive metastasis, and impair treatment outcomes. Senotherapeutics, have emerged as promising strategies for targeting senescent cells in cancer therapy. These agents selectively induce apoptosis in senescent cells while preserving normal tissues, representing a paradigm shift in oncology. Senotherapeutics can function as standalone treatments by clearing senescent tumor cells or as adjuvants to chemotherapy and radiotherapy, effectively eliminating residual therapy-induced senescent cells that may contribute to relapse. This dual approach allows for reduced treatment toxicity, improved therapeutic efficacy, and decreased tumor recurrence. Furthermore, targeting non-cancerous senescent cells may help suppress inflammation-driven tumorigenesis, slow disease progression, and enhance patient outcomes. Despite their promise, challenges remain in optimizing senotherapeutic strategies, identifying precise biomarkers, and minimizing off-target effects. This review explores the mechanisms of cellular senescence, its role in tumor dynamics, and the potential of senotherapeutics as a novel adjunct in cancer treatment. By integrating senotherapeutics with existing modalities, the field moves closer to more effective, personalized cancer interventions, warranting further preclinical and clinical investigation.
Read moreBusiness groups as knowledge-based hierarchies of firms
Abstract We provide the first worldwide overview of the patterns of hierarchical differentiation across Business Groups (BGs), highlighting the coexistence of different hierarchical shapes. We show how the different shapes can arise as optimal hierarchical structures in a knowledge-based model of BGs when subsidiaries’ operations involve problem-solving under parents’ supervision. The optimal choice of hierarchical structure is driven by production efficiency and two dimensions of problem-solving: efficiency related to supervising knowledge creation and handling associated communication across subsidiaries. We check the consistency of the model’s predictions with the empirical patterns. The model successfully passes the consistency test (JEL D23, L23, F23, L25, G34)
Read moreAstronomy: The initial conditions for planet formation
Astronomy: The initial conditions for planet formation Leonardo Testi, Ugo Lebreuilly, Elenia Pacetti, Anaëlle Maury, Veronica Roccatagliata, Patrick Hennebelle, Ralf Klessen, and Sergio Molinari investigate the initial conditions for planet formation in this special astronomy focus. Understanding the origins of our own Solar System and determining whether it is a common or rare occurrence in the cosmos is a fundamental question in modern astrophysics, as well as a topic of general interest to human society.
Read moreTelling the <i>gendered</i> story: the construction of migrant subjectivities in UNHCR’s animated information campaign
Abstract This article explores how the UNHCR exercises productive power through gendered knowledge in migration governance. Using multimodal semiotic analysis of twelve animated films from the Telling the Real Story campaign, we examine how public information campaigns (PICs) on irregular migration risks balance humanitarian protection and border control through gendered depictions. While PIC research is expanding, it often overlooks the role played by gender in the campaigns’ content. Our analysis shows that TRS’s animated films legitimize and reward female migrants’ agency only when aligned with UNHCR’s institutional goals, while pathologizing male mobility by consistently framing it as a threat or failure. These representations emerge across four themes: family (entrepreneurs vs caregivers), violence (aggressors vs victims), professional aspirations (money vs education) and autonomy (independent vs dependent). We argue that such discourse reinforces gender essentialisms, mainly through different renderings of migrant agency, legitimizing both humanitarian and securitizing migration governance while potentially undermining migrant empowerment.
Read moreExposing the Illusion: A Comprehensive Study on Fake Review Detection on Amazon
Nano-microfluidic platforms for the detection of environmental microbial pathogens
A Framework for Predicting Diabetes Using Machine Learning Techniques