Research works

Machine Learning Classification and Portfolio Construction: Does the Loss Function Matter?

Classification outperforms regression across matched machine learning models in portfolio construction. A stacking ensemble of gradient boosted trees, random forest, and neural network yields a value-weighted annualized Sharpe ratio of 2.08 for classification and 1.39 for regress...

Developing attributes and attribute-levels for a discrete choice experiment on preference for primary healthcare providers among urban slum dwellers in Bangladesh

Abstract Background Due to rapid urbanization, in many low- and middle-income countries (LMICs), urban primary healthcare (PHC) systems are not compatible with meeting the growing need for healthcare services. However, limited research exists on what patients consider important w...

Influence of weather variables on population dynamics of rice insect pests and their natural enemies: A six year light trap study in Gazipur, Bangladesh

Sustainable rice production in tropical and subtropical agroecosystems, it is essential to know, how climatic variability regulates the dynamics of insect pests and natural enemies. Our study evaluated the monthly and seasonal pattern of key weather variables and their influence...

Quantification of Urinary Free and Total Oxidative Stress Biomarkers and Their Differential Associations with Polycyclic Aromatic Hydrocarbon Exposure: A Bushfire Smoke Study

Abstract Urinary excretion profiles of free and metabolically conjugated oxidative stress biomarkers (OSBs) remain poorly characterized, potentially underestimating total excretion and biasing exposure–effect interpretation. In this study, the profiles of free, glucuronidated, an...

OUT-Guard: Context-Aware Risk Detection and Adaptive De-Identification of Rare and Extreme Information

Records containing rare and extreme information can pose substantial disclosure risks even after direct identifiers have been removed. Existing methods typically assess numerical outliers, rare categories, small contextual cells, and behavioral anomalies independently, limiting t...

Infrastructure as Inherited Knowledge: Artificial Intelligence and the Governance of Long‐Life Infrastructure

ABSTRACT Epistemic orphaning, responsibility drift, loss of institutional memory, and non‐stationarity are not new problems in infrastructure governance—they have accompanied long‐life infrastructure throughout its history. What artificial intelligence changes is not the existenc...

Impact of the 2023 ACR/EULAR classification criteria on thrombotic antiphospholipid syndrome: reclassification patterns and long-term outcomes in a 419-patient cohort

The 2023 ACR/EULAR classification criteria for antiphospholipid syndrome (APS) were developed to improve specificity and to define more homogeneous research populations. However, their performance in real-world cohorts of thrombotic APS with long-term hard outcomes remains insuff...

Global Existence and Exponential Stability of a Boundary Contact Problem for the 2D Mindlin–Timoshenko System

Abstract We analyze a contact problem for the two-dimensional Mindlin–Timoshenko system describing the vibratory motion of elastic plates in unilateral contact with a rigid obstacle along a portion of the boundary. Precisely, we establish the existence of solutions and prove that...

From Isolated to Complex: Diagnostic Yield of Chromosomal Microarray and Exome Sequencing in Fetal Growth Restriction

ABSTRACT Objective To characterize the overall diagnostic yield of sequential chromosomal microarray (CMA) and exome sequencing (ES) across the phenotypic spectrum of fetal growth restriction (FGR). Methods A retrospective cohort of 320 consecutive FGR pregnancies presenting at ≥...

Functional and Technological Characterization of Microorganisms Isolated from Wet Coffee Fermentation in Brazil: A Multivariate Approach for the Selection of Promising Isolates

Abstract Coffee fermentation is a stage of wet coffee processing carried out by microorganisms. This study aimed to characterize microorganisms isolated from wet coffee fermentation by their enzymatic potential and to evaluate how different stress conditions affect the selection...

Modeling Realistic Farmer Behavior in Heuristic-Based Irrigation Decisions with Fuzzy Logic

Abstract This study addresses the challenge of integrating complex human behavior into hydrological models by developing a behavioral simulation model (BSM) within a multiagent framework. Focusing on farmers’ response to water scarcity in Iran’s Lake Urmia basin, the BSM utilizes...

Biomechanical analysis of lower limb asymmetry in elite junior male Tai Chi athletes: associations with stomping technique performance

Abstract Long-term unilateral dominant training in Tai Chi is commonly associated with lower-limb functional asymmetry in elite athletes, which may correlate with reduced movement proficiency and increased injury susceptibility. This study recruited 80 elite junior male Tai Chi a...

Financial, innovation and fiscal drivers of ecological efficiency across Western and Eastern European Union member states

Abstract This study quantifies how green finance, financial development, green technology innovation efficiency and environmental taxes shape ecological efficiency in the European Union, and tests whether these effects differ between Western and Eastern member states. The sample...

Embryos resolve the enigmatic origin of the theropod wishbone

The origin of the avian wishbone (furcula)—present since theropod dinosaurs—remains unsolved. Its homology is still debated, with competing hypotheses proposing derivation from the clavicles, the interclavicle, or both. To obtain comparative developmental evidence, we studied oss...

Intraoperative tachycardia burden, diastolic hypotension, and in-hospital mortality after noncardiac surgery: a retrospective cohort study

Intraoperative hypotension is an established perioperative risk marker, but the relevance of intraoperative tachycardia remains uncertain. We examined tachycardia, diastolic blood pressure (DBP), mean arterial pressure (MAP), and temporal co-occurrence in noncardiac surgery. We c...

Genome accessibility reveals disease risk that gene expression misses

Most disease-associated genetic alterations do not affect the genes themselves, but instead change how strongly they are expressed, often by modifying genome accessibility. Gene expression and accessibility have now been analysed simultaneously in 10 million immune cells from 1,1...

Exploring health anxiety among healthcare professionals: a scoping review and exploratory quantitative summary of health anxiety findings

Health anxiety (HA) is characterised by an excessive fear or preoccupation with illness, often accompanied by a persistent belief in having a serious condition, despite medical reassurance. Among those affected, HA can significantly impact mental and physical wellbeing. While HA...

Algorithmic Visibility and Automated Continuation: Infrastructural Mediation in Streaming Platforms

AI-mediated communication is commonly associated with conversational agents, generative systems, and emotionally responsive interfaces. This article examines Top-10 rankings and autoplay as forms of infrastructural mediation embedded in streaming-platform interfaces and discusses...

Corporate governance and HR ethics: case insights on compliance programs

Abstract This study examines how compliance programs operate as governance mechanisms, shaping the ethical conduct of human resources over time. Drawing on Michel Foucault’s conceptualization of ethics, the paper analyses how different configurations of compliance practices influ...

Continuous spatiotemporal variations of riverine dissolved organic carbon in the Yangtze River Basin revealed by a river-reach-scale machine learning-based model

Study region The Yangtze River Basin (YRB), China Study focus Riverine dissolved organic carbon (DOC) is critical to the global carbon cycle, while its continuous spatiotemporal variations remain poorly understood in large rivers. This study developed an estimation model at the r...