High-frequency banking environments face a critical trade-off between low-latency fraud detection and the regulatory explainability demanded by GDPR. Traditional rule-based and discriminative models struggle with "zero-day" attacks due to extreme class imbalance and the lack of historical precedents. This paper proposes a Dual-Path Generative Framework that decouples real-time anomaly detection from offline adversarial training. The architecture employs a Variational Autoencoder (VAE) to establish a legitimate transaction manifold based on reconstruction error, ensuring <50ms inference latency. In parallel, an asynchronous Wasserstein GAN with Gradient Penalty (WGAN-GP) synthesizes high-entropy fraudulent scenarios to stress-test the detection boundaries. Crucially, to address the non-differentiability of discrete banking data (e.g., Merchant Category Codes), we integrate a Gumbel-Softmax estimator. Furthermore, we introduce a trigger-based explainability mechanism where SHAP (Shapley Additive Explanations) is activated only for high-uncertainty transactions, reconciling the computational cost of XAI with real-time throughput requirements.
This study explores the challenges and implications of Corporate Social Responsibility (CSR) in the banking sector of Bangladesh, highlighting its regulatory framework, implementation gaps, and alignment with sustainable development goals. While the central bank mandates CSR, the profit-driven nature of banking institutions often shifts the focus of CSR initiatives toward competitive advantage and brand enhancement rather than addressing genuine social and environmental needs. Major investments are concentrated in the education and health sectors, with minimal attention to ecological sustainability and marginalized communities. Weak regulatory oversight, profit-oriented governance structures, and limited stakeholder participation hinder the effective implementation of CSR. The lack of diversity in board representation, particularly the exclusion of women and underrepresented groups, further limits CSR's participatory and inclusive nature. This study underscores the need for stronger policy interventions, enhanced monitoring mechanisms, and a shift in corporate governance to transform CSR into a tool for meaningful societal impact. The findings call for further research to explore strategies for aligning profit-driven motives with sustainable and equitable development objectives.
This paper studies the influence of behavioral biases on Fintech adoption. Additionally, the role of financial literacy in adaptation of Fintech services is evaluated. Primary data from customers in the banking sector is gathered using a structured questionnaire. Factor analysis, reliability analysis, and correlation analysis are performed as diagnostic tests. The core purpose is examined through the application of regression model and moderation analysis. Findings reveal that risk-averse customers and overconfident customers are less likely to adopt Fintech services. Additionally, anchoring bias and disposition bias also hinder customers from using Fintech services offered by the banks. Overall, the findings suggest that psychological biases can hinder customers' choices of Fintech services. Financial literacy seems to improve the relationship in a positive direction as people who are financially literate are more adaptive towards Fintech services. The findings of this paper have strong implications for academicians and practitioners. Academicians can find a direction for imparting financial literacy to enhance the acceptability of Fintech services. Also, the findings imply that practitioners should design user friendly interfaces and enhance financial literacy efforts to mitigate bias-driven resistance and improve adoption rates. The study links behavioral biases with the developing sector of Fintech services. Traditionally, investment decisions as an outcome of behavioral biases were the focused dimension. This study opens avenues for research in modern realms of Artificial Intelligence and Fintech.
Futures, as significant financial derivatives, play a crucial role in financial markets by fulfilling price discovery functions and providing efficient risk hedging tools. Against the backdrop of geopolitical conflicts, market risk emerges not only from external shocks and random fluctuations but also from strategic interactions among diverse participants including hedgers, speculators, arbitrageurs, and regulators. This study integrates traditional VaR theory with machine learning methods to systematically examine risk characteristics and transmission mechanisms in the sugar futures market under geopolitical uncertainty. Utilizing sugar No. 5 futures trading data from the Zhengzhou Futures Exchange spanning 2015–2019 and 2024, we employ a Random Forest model for feature importance analysis and compare three risk measurement approaches: traditional parametric VaR, historical simulation methods, and machine learning-enhanced VaR models. We conduct empirical tests to validate the theoretical relationship √3 × VaRT(1,p) ≈ VaRT(3,p) and calculate epsilon values (relative deviation between actual and estimated tail risk occurrences) through return tests. Annual delta values range between 0.26 and 1.16, averaging approximately 35% below theoretical values. The machine learning-based Value at Risk (VaR) at 95% confidence level exhibits a violation rate of 5.00%, demonstrating superior accuracy compared to parametric VaR (26.67%) and traditional historical VaR (7.00%). Epsilon values show no statistically significant difference between 2024 (0.08) and the 2015–2019 average level (0.14), indicating stable risk transmission mechanisms despite geopolitical conflicts. The hybrid “machine learning-traditional theory” risk framework developed in this research provides a theoretical foundation and practical guidance for regulatory bodies to enhance risk prevention and control systems, as well as for market participants to optimize risk management strategies. Despite geopolitical impacts, the fundamental risk transmission mechanisms of the sugar futures market remain relatively stable, demonstrating market resilience.
Tetyana Ivanova, Halyna Kryshtal, Volodymyr Metelytsia
et al.
This article analyzes the implementation of investment-oriented innovations in the field of medical insurance to establish an effective and sustainable healthcare financing model amidst digital economic transformation. The authors provide theoretical justification for the role of innovations and investments in the development of the medical insurance sector, along with a deep analysis of the current state of medical services and the main barriers hindering innovation in this area. Special attention is given to assessing the impact of digitalization on the accessibility and quality of medical services, as well as the potential of using advanced technologies such as telemedicine and online platforms for insurance companies. It is considered that digital technologies, including telemedicine and online services, have the potential to significantly improve access to medical services and reduce costs. However, the implementation of these innovations faces several challenges, including issues of internet access, the lack of proper legal frameworks for regulating digital technologies in medical insurance, and the need to enhance financial literacy among the population. The authors summarize and propose a list of barriers (technical, regulatory, financial, cultural, psychological barriers, barriers related to human resources, and competition-related barriers) in attracting investments to the sector. Practical recommendations are offered for stakeholders, including insurance companies, government bodies, and users of medical services, to facilitate more effective innovation implementation and ensure sustainable financing of medical services. The strategic approaches to sustainable healthcare financing proposed in the article aim to ensure equal access to medical services for all population segments, improve the quality of insurance services, and optimize costs through digital tools.
While individuals in emerging markets are concerned about climate change, such concerns do not necessarily translate into a willingness to pay for environmental policies. Using rich data for 37 economies, mostly from emerging markets in Europe, the Caucasus, Central Asia and parts of North Africa and the Middle East, we empirically examine correlations with willingness to pay for environmental policies. We show that, beyond ability to pay, people who expect to be better off in the future, who are more patient and who trust the government are all more likely to be willing to pay for policies that mitigate climate change. Our results thus suggest that measures that increase people’s incomes, build trust in government, reduce corruption and increase the transparency and efficiency of government spending could help boost support for green policies. Policies may also receive greater support if they take the form of subsidies, where the costs in terms of higher taxes are less salient.
The research aims to explain the procedures that banks take to combat money laundering, in addition to identifying the obstacles that hinder banks from carrying out their duties towards combating money laundering. The research was divided into three sections. The first section dealt with the concept of money laundering, and identifying the stages of the money laundering process. The sources of money laundering operations were clarified, and the second section dealt with banks’ procedures to reduce money laundering operations. In the third section, the responsibility of banks towards detecting money laundering operations was clarified, through reviewing the control system for money laundering operations, as well as the preventive measures required by banks. To combat money laundering, in addition to the mechanism for reporting money laundering operations, the research found that, through money laundering operations, several illegal and illegal activities are carried out, resulting in funds in very large quantities, which negatively affects the country’s economy.
Purpose – This inquiry aims to examine the impact of Green Banking performance and service quality on customer satisfaction using Mobile Banking. Methodology - In this inquiry using the type of primary data by survey method, which makes a questionnaire with questions related to research. The data analysis technique used is multiple linear regression analysis. This technique is useful to find out the impact of several independent elements on the dependent element analyzed through SPSS version 29. Findings – Based on the results obtained that the performance of green banking and service quality significantly affect customer satisfaction. Together, green banking performance and service quality have a significant influence on customer satisfaction using mobile banking. Implications – The results of this inquiry will provide recommendations on the association between the performance of green banking and mobile banking is increasingly crucial in the context of digital transformation in the banking sector that focuses on sustainability. Originality – This inquiry provides insights to increase attention to the concept of sustainability in the modern banking world, especially in the application of green banking.
Abderrahman Skiredj, Ferdaous Azhari, Ismail Berrada
et al.
Navigating the complexities of language diversity is a central challenge in developing robust natural language processing systems, especially in specialized domains like banking. The Moroccan Dialect (Darija) serves as the common language that blends cultural complexities, historical impacts, and regional differences. The complexities of Darija present a special set of challenges for language models, as it differs from Modern Standard Arabic with strong influence from French, Spanish, and Tamazight, it requires a specific approach for effective communication. To tackle these challenges, this paper introduces \textbf{DarijaBanking}, a novel Darija dataset aimed at enhancing intent classification in the banking domain, addressing the critical need for automatic banking systems (e.g., chatbots) that communicate in the native language of Moroccan clients. DarijaBanking comprises over 1,800 parallel high-quality queries in Darija, Modern Standard Arabic (MSA), English, and French, organized into 24 intent classes. We experimented with various intent classification methods, including full fine-tuning of monolingual and multilingual models, zero-shot learning, retrieval-based approaches, and Large Language Model prompting. One of the main contributions of this work is BERTouch, our BERT-based language model for intent classification in Darija. BERTouch achieved F1-scores of 0.98 for Darija and 0.96 for MSA on DarijaBanking, outperforming the state-of-the-art alternatives including GPT-4 showcasing its effectiveness in the targeted application.
The 3D flight control of a flapping wing robot is a very challenging problem. The robot stabilizes and controls its pose through the aerodynamic forces acting on the wing membrane which has complex dynamics and it is difficult to develop a control method to interact with such a complex system. Bats, in particular, are capable of performing highly agile aerial maneuvers such as tight banking and bounding flight solely using their highly flexible wings. In this work, we develop a control method for a bio-inspired bat robot, the Aerobat, using small low-powered actuators to manipulate the flapping gait and the resulting aerodynamic forces. We implemented a controller based on collocation approach to track a desired roll and perform a banking maneuver to be used in a trajectory tracking controller. This controller is implemented in a simulation to show its performance and feasibility.
Some literature suggests that Islamic banking, particularly in terms of financing agreement, always employs standard contracts. The contracts are agreement to the clauses that have already been established unilaterally by one party possessing a stronger position; the party is the bank. Consequently, customers basically cannot negotiate for the content of the contracts. It is worth mentioning that the unilateral decision is contrary to the Surah An-Nisa Verse 29 asserting that trade should be based on a consensus between two parties. The research findings reveal that the standard contracts can be justified in the view of Islam since it is a ‘mubah’ case that is not prohibited by the Qur’an in relation to ‘qath’y’, especially the ‘maqasid’ (purposes) of the standard contracts for the sake of ease and acceleration in the contracts. For instance, the contract No. 007/WKL/UMS/0117/9310/IV/2013 BRI Syariah is actually in accordance with sharia principles. This contract discusses the time for purchasing goods dealing with ‘murabahah’ (cost plus profit) financing, but the implementation of the contract is deemed inappropriate according to the sharia principles.
Faraz Nabiyi , Mehdi Shamizanjani, Nima Garoosi Mokhtarzadeh
Today's world has been affected by digital technologies more than ever, and no business can be considered independent of these technologies, regardless of industry, size, and geography. Various industries such as banking, insurance, automobile, petrochemical, steel, energy, food, entertainment and education are experiencing a fundamental change centered on digital technologies. In the current era, organizations have entered a new era in which digital technologies have revolutionized the majority of industries, from products and services to customer expectations. Today's organizations have practically no other way to survive and that is digital transformation. Digital transformation is an inevitable necessity and a vital strategic issue for today's businesses, which will suffer great losses if they do not adopt and implement the right strategy and consequently they might fail.By reviewing the theoretical academic resources, despite the need for a model to formulate a digital transformation strategy which there is a consensus on and contains various dimensions of digital transformation with an exhaustive view, there has never been such a framework, which reminds us of the need to address this issue. This research is designed to address the Conceptualization of digital transformation strategy as mentioned earlier. In this research, by conceptualizing the digital transformation strategy and cognition of its various dimensions, the mentioned gap will be addressed, by using the research method of meta synthesis. For this purpose, after reviewing 17 selected articles, a specific definition of digital transformation strategy has been presented and its relationship with business strategy and functional strategies has been analyzed. Also, the principles of shaping the digital transformation strategy are presented in two content and process categories. Finally, three categories of cultural, structural and leadership capabilities have been identified as supporting capabilities for shaping the digital transformation strategy.
This paper presents the results of an assessment of the impact of country and ownership specific features of banks on the results of the EBA cyclical stress tests. For the purpose of panel model estimation a dedicated database has been built by the author. The results of the random effects panel model support some anecdotal evidence on the influence of local regulators. The model results also prove the efficiency of restructuring plans. They helped banks with the plan to better prepare for turbulent times. Ownership structure had a significant impact on stress-test performance. Being a bank with the state as the largest shareholder or being a cooperative bank significantly improved the performance of the bank in the EBA stress tests. The research focuses on qualitative characteristics assuming no impact of banks' finances or business cycle.
The food landscape of Calgary, Canada, is sown with an abundance of polycultures. Alongside place-specific Indigenous foodways are food rescue, banking, and hamper programs, food studies scholars, a City of Calgary food resilience plan, and a growing number of alternative food network producers. Within the local alternative food network, there has been a boom in advancing indoor growing for our colder climate, including container, aquaponic, vertical hydroponic, and greenhouse growing. Situated as an agrarian ethnographer and an urban regenerative farmer, we seek to highlight the viability of agricultural techniques that are in relation with the land to grow more socially and ecologically sustainable food and farm systems in and around Calgary. From this position, we formed a collaboration between the University of Calgary, Root and Regenerate Urban Farms, and the Young Agrarians to document the cultivation process for a production urban farm. Over the course of one growing season—May to September, 2021—we harvested approximately 7,000 lbs (3,175 kg) of produce across nine urban spaces totaling 0.26 acres. The 48 vegetable varieties were distributed to 35 community supported agriculture shareholders, weekly farmers market customers, restaurant chefs, and members of the YYC Growers and Distributors cooperative. Moreover, we donated 765 lbs (347 kg) of surplus produce to the Calgary Community Fridge, Calgary Food Bank, and the Alex Community Food Centre, which work to mitigate food insecurity. Through a reflexive practitioner approach, our reflective essay discusses the benefits and limitations of Small Plot Intensive Farming methods and urban land-sharing strategies, as well as the viability of land-based urban agriculture in a rapidly changing socio-ecological climate. Our paper also demonstrates the potential for transcending siloed approaches to knowledge-making vis-à-vis experiential learning partnerships between graduate student researchers, farmers, and agricultural organizations.
Nghiên cứu có mục đích là xác định các thuộc tính của kênh bán lẻ trực tuyến là website thương mại điện tử và sàn thương mại điện tử như là Shopee, Lazada, Tiki, Sendo, … ở Việt Nam. Phương pháp nghiên cứu hỗn hợp với 18 người mua sắm được phỏng vấn để khám phá các thuộc tính và 288 người mua sắm được khảo sát. Phương pháp phân tích là phân tích nhân tố khám phá (EFA) được sử dụng để nhóm các thuộc tính và phân tích nhân tố khẳng định (CFA) được sử dụng để kiểm tra mô hình. Nghiên cứu xác định 04 thuộc tính bán lẻ trực tuyến là: Dịch vụ giao hàng, Tiện lợi không gian và thời gian, Dễ dàng lựa chọn hàng hóa và Hàng hóa.
Lenka Moravcová, Angelino Carta, Petr Pyšek
et al.
Abstract Soil seed viability and germinability dynamics can have a major influence on the establishment and spread of plants introduced beyond their native distribution range. Yet, we lack information on how temporal variability in these traits could affect the invasion process. To address this issue, we conducted an 8-year seed burial experiment examining seed viability and germinability dynamics for 21 invasive and 38 naturalized herbs in the Czech Republic. Seeds of most naturalized and invasive species persisted in the soil for several years. However, naturalized herbs exhibited greater seed longevity, on average, than invasive ones. Phylogenetic logistic models showed that seed viability (but not germinability) dynamics were significantly related to the invasion status of the study species. Seed viability declined earlier and more sharply in invasive species, and the probability of finding viable seeds of invasive species by the end of the experiment was low. Our findings suggest that invasive herbs might take advantage of high seed viability in the years immediately after dispersal, while naturalized species benefit from extended seed viability over time. These differences, however, are not sufficiently strong to explain the invasiveness of the species examined.
Establishing a new business may involve Knowledge acquisition in various areas, from personal to business and marketing sources. This task is challenging as it requires examining various data islands to uncover hidden patterns and unknown correlations such as purchasing behavior, consumer buying signals, and demographic and socioeconomic attributes of different locations. This paper introduces a novel framework for extracting and identifying important features from banking and non-banking data sources to address this challenge. We present an attention-based supervised feature selection approach to select important and relevant features which contribute most to the customer's query regarding establishing a new business. We report on the experiment conducted on an openly available dataset created from Kaggle and the UCI machine learning repositories.
When searching for new gravitational-wave or electromagnetic sources, the $n$ signal parameters (masses, sky location, frequencies,...) are unknown. In practice, one hunts for signals at a discrete set of points in parameter space, with a computational cost that is proportional to the number of these points. If that is fixed, the question arises, where should the points be placed in parameter space? The current literature advocates selecting the set of points (called a "template bank") whose Wigner-Seitz (also called Voronoï) cells have the smallest covering radius ($\equiv$ smallest maximal mismatch). Mathematically, such a template bank is said to have "minimum thickness". Here, for realistic populations of signal sources, we compute the fraction of potential detections which are "lost" because the template bank is discrete. We show that at fixed computational cost, the minimum thickness template bank does not maximize the expected number of detections. Instead, the most detections are obtained for a bank which minimizes a particular functional of the mismatch. For closely spaced templates, the fraction of lost detections is proportional to a scale-invariant "quantizer constant" G, which measures the average squared distance from the nearest template, i.e., the average expected mismatch. This provides a straightforward way to characterize and compare the effectiveness of different template banks. The template bank which minimizes G is mathematically called the "optimal quantizer", and maximizes the expected number of detections. We review optimal quantizer and minimum thickness template banks that are built as n-dimensional lattices, showing that even the best of these offer only a marginal advantage over template banks based on the humble cubic lattice.
Alessandro Castelnovo, Riccardo Crupi, Giulia Del Gamba
et al.
Algorithmic bias mitigation has been one of the most difficult conundrums for the data science community and Machine Learning (ML) experts. Over several years, there have appeared enormous efforts in the field of fairness in ML. Despite the progress toward identifying biases and designing fair algorithms, translating them into the industry remains a major challenge. In this paper, we present the initial results of an industrial open innovation project in the banking sector: we propose a general roadmap for fairness in ML and the implementation of a toolkit called BeFair that helps to identify and mitigate bias. Results show that training a model without explicit constraints may lead to bias exacerbation in the predictions.
This paper considers the problem faced by a bank which trades in the funds market so as to maintain the reserve requirements and minimize the costs of doing that. We work in a stochastic paradigm and the reserve requirements are determined by the demand deposit process, modelled as a geometric Brownian motion. The discount rates for the cumulative funds purchased and the cumulative funds sold are assumed to be different. The optimal strategy of the bank is explicitly found and it has the following structure: when bank reserves lower to an exogenously threshold level the bank has to purchase funds; when bank reserves tops an endogenously threshold level the bank has to sell funds