Vivekanand Education Society Institute of Management Studies & Research (VESIM) is a premier business school located in Mumbai, Maharastra, India.
Advanced detection systems are increasingly necessary as digital financial fraud grows alongside the rapid expansion of Unified Payments Interface (UPI) transactions. This study presents a hybrid fraud detection framework built on a two-path architecture that combines supervised machine learning with unsupervised anomaly detection. The model is trained and analyzed on the PaySim dataset, containing 6.35 million synthetic mobile money transactions with a 10% enhanced SMOTE generated fraud rate. Path A focuses on known fraud patterns using XGBoost and Random Forest classifiers trained on multiple engineered features, including Autoencoder reconstruction error, enabling automatic fraud blocking with 0.996 recall, 0.998 precision, 0.998 accuracy, an F1-score of 0.998, and an ROC-AUC of 0.996. The Autoencoder is optimized using early stopping over up to 40 epochs, achieving optimal convergence at epoch 35 with a validation loss of 2.53×10−1. Path B identifies previously unseen fraud using a standalone Autoencoder and Isolation Forest, detecting about 0.90 of new fraud cases for human review. Overall, the hybrid approach outperforms single models and OR-based ensembles, reducing false positives while maintaining strong detection of evolving fraud patterns.
The quest for quality in management education requires urgent attention with the emergence of Generation Z workforce. The literature falls short of providing a comprehensive scale for quality management education. The authors adopted a mixed-method approach for conducting the study. Fifteen qualitative interviews with faculty, students, and industry professionals were conducted, and helped to craft 49 questions in final questionnaire. Eight factors extracted using principal component analysis, subsequently these factors were clubbed into three dimensions. This novel study promises to expand our understanding by offering multiple implications for addressing SDG goals and providing valuable insights for future research and policymaking.
Smart cities have more cyber-physical production systems and interconnected urban logistics involved. With the use of the survival-function model of the operation survey approach presented in this work, the logistics information that is generated by smart manufacturing reliability maintenance (predictive maintenance) is integrated with more elevated performance parameters for large-scale operational resilience. Toward this goal, using the AI4I 2020 predictive maintenance dataset, which consists of 100,000 industrial data, we implemented two supervised learning approaches (balanced Logistic Regression and Random Forest) and established their ability to prevent machines from failing. Our outcomes, in particular, show that the Random Forest model had a ROC-AUC equal to 0.9895, a high value for such random search approaches for rare event detection. This highlights its effectiveness in predicting rare failure modes. We also consider how such models facilitate condition-based maintenance, increase operational efficiency by minimizing unexpected downtimes, and increase the safety of last-mile as well as plant-to-plant logistics through logistical trade-off enhancements in terms of routing and resource capacity. We additionally include quantitative verification (ROC-AUC, PR-AUC, confusion matrices), significance of the features, and the way forward that can be applied in smart city projects of a huge scale.
Management education faces increased pressure to prepare students for workplaces characterised by technological disruption, changing skill requirements, and heightened expectations of immediate job readiness. While existing research has examined Generation Z's employability, digital competencies, and expectations from higher education, less is known about how management students themselves perceive the process through which classroom knowledge becomes career-relevant capability. This study explores how Generation Z postgraduate management students conceptualise career readiness, identify perceived gaps between management education and workplace expectations, and describe the learning experiences they believe facilitate the transition from knowing to doing. Guided by an interpretivist paradigm, the study adopts a participatory netnographic design to explore how Generation Z postgraduate management students construct meanings of career readiness through online community interactions. Data is collected through student discussions, reflective posts, polls, and researcher-facilitated conversations within a dedicated digital community and analysed using reflexive thematic analysis. The study aims to develop a Knowledge-to-Capability Translation Framework that explains how students perceive the movement from conceptual learning to workplace readiness through experiential learning, digital competence, AI literacy, professional confidence, and capability development.
The Government of India runs many welfare schemes to support people from economically weaker sections of society. These schemes are meant to help citizens by providing financial assistance, but in reality, the current fund distribution process often faces problems such as delays, corruption, lack of transparency, and difficulty in tracking where the money actually goes. Because of these issues, benefits sometimes do not reach the right people at the right time. To solve this problem, this project proposes an improved E-Fund Distribution System using Blockchain technology and Smart Contracts. In this system, funds are automatically transferred when a beneficiary meets the required eligibility conditions, which reduces manual work and removes unnecessary intermediaries. Every transaction is recorded on a secure and decentralized blockchain network, making it transparent and impossible to modify. This ensures better accountability and proper monitoring of public funds. By combining blockchain with the existing digital governance system, the proposed solution aims to make the welfare fund distribution process more reliable, transparent, and efficient across India.