Westcliff University

Abstract Selecting an appropriate supportive response for a mental health question post is a key step toward scalable mental health question answering, yet it remains difficult due to semantic mismatch, noisy informal language, and the need to align supportive intent beyond surface lexical overlap. We propose MF-GAT, a novel multi view graph attention matching framework that constructs a Concept Interaction Graph (CIG) to explicitly encode post response concept alignments and their interaction structure. MF-GAT learns three complementary evidence streams, including local interaction features, multi view fusion features, and global context features, and applies view specific graph attention to propagate and reweight informative relational signals over the CIG. A gated fusion module then adaptively integrates the view representations into a unified matching vector for prediction. We evaluate MF-GAT both as a pair classification model and as a retrieval ranking model for selecting the best support from a candidate pool, reporting Accuracy and F1 together with standard ranking metrics including MRR and NDCG. On the MHQA benchmark, MF-GAT achieves 0.95 Accuracy and 0.85 F1, outperforming BERT (0.89, 0.67), CIG-GCN (0.92, 0.79), ARC-II (0.85, 0.60), and MatchPyramid (0.82, 0.59). These results show that novel multi view interaction graph modeling with attention based propagation improves both supportive response classification and practical retrieval quality for mental health support selection.

Background: In an increasingly volatile global business environment, supply chain resilience has become a strategic imperative, particularly for firms operating in developing economies. Guided by Dynamic Capabilities Theory (DCT), this study examines how AI-powered tools foster an innovation culture comprising communication, creativity, and learning, and how these dimensions enhance supply chain resilience measured through flexibility, efficiency, and velocity. Methods: A quantitative research design was employed using survey data collected from 270 supply chain and operations managers in Jordanian manufacturing firms. Twelve direct hypotheses were tested using Partial Least Squares Structural Equation Modeling. Results: The findings indicate that AI-powered tools significantly influence communication, creativity, and learning. Communication and creativity positively affect all three dimensions of supply chain resilience. Learning significantly improves efficiency but shows no significant effect on flexibility or velocity, indicating that learning is mainly utilized for process improvement rather than rapid adaptation. Conclusions: The study demonstrates that AI adoption alone is insufficient to build resilient supply chains unless supported by innovation-oriented cultural capabilities. The findings extend DCT by clarifying the differentiated role of learning in resilience building and provide actionable guidance for managers seeking to align AI investments with cultural development in resource-constrained manufacturing contexts and long-term competitive advantage.

U.S. financial institutions deploying AI-based fraud detection face a fragmented compliance landscape spanning four regulatory frameworks – OCC Bulletin 2011-12, SR 11-7, the CFPB AI circular, and FinCEN BSA/SAR requirements – with no integrated governance life cycle connecting these requirements to model development, validation, and monitoring practice. This paper presents the Regulatory Governance Framework for AI-Driven Financial Fraud Detection (RGF-AFFD), a three-tier governance architecture empirically anchored in a multi-study empirical program. Using the IEEE-CIS dataset (590,540 transactions) and ULB benchmark (284,807 transactions), we benchmark six architectures including an LSTM+XGBoost ensemble, and conduct ablation, temporal drift, SHAP interpretability, and BISG fairness analyses. The LSTM+XGBoost ensemble achieves ROC-AUC of 0.9289 (F1: 0.6360) with a benefit-cost ratio of 6:1. XGBoost demonstrates the strongest temporal stability (delta-AUC = -0.0017 versus -0.0626 for LSTM). The RDT-FG Regulatory Digital Twin meta-model translates metrics into four regulator-specific health scores and a composite Regulatory Fitness Index for continuous compliance monitoring. The RGF-AFFD is the first integrated deployment blueprint to simultaneously satisfy OCC, SR 11-7, CFPB, and FinCEN requirements, supported by a community bank implementation vignette and four evidence-based policy recommendations.

BACKGROUND:Artificial intelligence (AI) is increasingly embedded in healthcare businesses, promoted for its ability to enhance efficiency, reduce costs and optimize workflows. However, the intersection of profit-driven priorities with patient-centred values presents significant ethical and professional challenges for nurses, who serve as the frontline mediators between technology and patients. AIM:This study aimed to explore nurses lived experiences of AI integration in healthcare businesses, focusing on how they navigate tensions between institutional efficiency and their professional commitment to patient-centred care. METHODS:An interpretive phenomenological design was employed to capture the depth of nurses' perspectives. Data were collected between May and June 2025 through 26 semi-structured interviews and 1 focus group with 7 nurses, yielding a total of 33 participants from AI-integrated private hospitals. Transcripts were analyzed thematically, with trustworthiness ensured through member validation, audit trails and reflexive journaling. RESULTS:Four overarching themes emerged. Nurses reported emotional and ethical conflicts when AI recommendations contradicted clinical judgement, often leading to moral distress. Business imperatives were perceived to prioritize efficiency over individualized care, with nurses excluded from decision-making about AI adoption. Many participants expressed anxiety over role displacement and a diminishing sense of autonomy, although some redefined their professional identity as technology navigators. Inadequate training and lack of institutional support further amplified challenges, leaving nurses underprepared to manage AI tools effectively. CONCLUSION:While AI offers organizational advantages, its integration without inclusive planning and adequate training risks undermining holistic nursing practice. Strengthening institutional support, valuing nurses' input and balancing efficiency with empathy are essential to align technological innovation with compassionate, patient-centred care.

Plasmodium parasites are the cause of malaria, a deadly illness that continues to pose a serious danger to world health, especially in areas with low resources where subjectivity, complexity, along with privacy issues make it difficult to employ traditional diagnostic techniques like microscopy and quick diagnostic testing. To overcome these specific challenges of diagnostic subjectivity, logistical complexity, and data privacy, this paper suggests a privacy-preserving federated learning system that uses sophisticated Vision Transformers (ViTs) for automated malaria identification from blood smear images. This paper suggests a privacy-preserving federated learning system that uses sophisticated Vision Transformers (ViTs) for automated malaria identification from segmented red blood cell (RBC) images in order to get around these issues. This architecture successfully addresses important privacy and logistical restrictions by enabling cooperative training among decentralized institutions without exchanging sensitive data. Prominent centralized convolutional neural networks (CNNs) are matched in diagnostic accuracy by the federated ViT models, which include ViT-B/16, DeiT-Tiny, Swin-T, and DINOv2. Interestingly, the federated transformer variations outperform the CNN ensemble (ResNet50 + VGG16) with an accuracy of 98.15%, FedDistill-DeiT achieving 97.79%, FedAvg-Swin-T reaching 97.75%, and FedDistill-Swin-T achieving a high ROC-AUC of 0.9977. These findings show that, even in the presence of diverse data distributions, federated Vision Transformers provide a reliable, scalable, and interpretable malaria screening solution that combines high accuracy with solid privacy guarantees.