Credit Cards Research Papers - Academia.edu (original) (raw)

2025, Deep Science Publishing

As long as financial transactions are recorded and exchanged, it has always been unavoidable that irresponsible actors, or risk factors, engage in various acts of fraud to gain wealth unethically. Such unauthorized transactions include stealing of payment method information, creating unauthorized accounts for receiving gain from illegal activities such as money laundering, creating unauthorized merchant accounts to trick customers into fraudulent transactions, and phishing to swindle customers. Payment processing companies should detect such behaviors to protect legitimate users and save their companies from losing a considerable amount of money (Ghemawat et al., 2003; Kimball & Ross, 2013; Akidau et al., 2015). Growing up in an internet-enriched environment, millennials and Generation Z are the largest active user segment. They are highly sensitive to payment transaction efficiencies and prone to try new payment methods compared to Generation X and Baby Boomers. These new trends require payment service providers to design novel transaction processing systems to accommodate different payment methods, account users, and merchants, while keeping real-time fraud detection at tolerable costs. Acquiring, storing, and processing the intensive data streams generated by users during payment transactions are core challenges for payment processing services. Streaming data engineering is the cornerstone for building such payment processing and risk analysis pipelines. Data such as transactions, partner systems, and payment gateways are collected to a data lake from internal and external sources. Thereafter, the collected data are processed using batch or streaming pipelines to provide real-time transactional and risk insights for the payment operations organizations and partners. Those insights are then ingested into dashboards for tracking transaction activity and risk detection performance (Sadoghi & Jacobsen, 2011; Zhang & Xu, 2020).

2025, Deep Science Publishing

2025, IEEE open journal of signal processing

The creation of the dataset has been supported by Deutsche Gesellschaft f ür Internationale Zusammenarbeit (GIZ) on behalf of the German Ministry for Economic Cooperation and Development.

2025, International Journal of Computer Science and Mobile Computing (IJCSMC)

With raising in-depth amalgamation of the Internet and social life, the Internet is looking differently at how people are learning and working, meanwhile opening us to growing serious security attacks. The ways to recognize various network threats, specifically attacks not seen before, is a primary issue that needs to be looked into immediately. The aim of phishing site URLs is to collect the private information like user's identity, passwords and online money related exchanges. Phishers use the sites which are visibly and semantically like those of authentic websites. Since the majority of the clients go online to get to the administrations given by the government and money related organizations, there has been a vital increment in phishing threats and attacks since some years.

2025, International Journal of Computer Science and Mobile Computing (IJCSMC)

Identifying early signs of corporate distress remains a complex challenge, particularly in today’s unpredictable economic landscape. In this study, we examined whether sentiment analysis and textual data could enhance the prediction of impending corporate closures. Data was collected from 30 companies across various industries, including employee reviews, financial press releases, and other publicly available financial information.
Sentiment scores were calculated using established tools like TextBlob and VADER. Through keyword frequency analysis, we identified recurring negative themes—terms such as “layoffs,” “mismanagement,” and “low morale” appeared frequently among companies that ultimately ceased operations. We applied Principal Component Analysis (PCA) to reduce dimensionality, then utilized classification algorithms, specifically Random Forest and K-Nearest Neighbors, to predict closure outcomes.
Our results indicate that models incorporating sentiment and textual features significantly outperformed those relying solely on traditional financial indicators. Notably, the Random Forest model achieved the highest accuracy and F1-score.
These findings underscore the value of integrating qualitative employee insights with conventional risk metrics. This approach offers a more comprehensive and proactive framework for forecasting organizational failure, with important implications for investors, HR professionals, and policymakers seeking to implement effective early warning systems.

2025, ijbssnet.com

Islamic banking products and services are gaining popularity among non-Muslims across the globe due to its wider product coverage and ability to traverse the global economic melt-down. However, to what extent this statement is true in... more

2025, ACR North American Advances

In this paper we show that reducing the subjective value of money decreases sensitivity to losses. We find that using less tangible forms of money increases the amount of money gambled, the propensity of selecting riskier gambles, and... more

2025

2025, International Journal of Management Research and Business Strategy

To improve fraud detection in the banking industry, the study investigates integrating neural networks with the Harmony Search Algorithm (HSA). Advanced solutions are required since traditional procedures are frequently insufficient in the face of sophisticated fraud techniques. HSA is combined with neural networks, which are renowned for their capacity to recognize intricate patterns, to effectively optimize the settings. This study shows that fraud detection accuracy and reliability are greatly increased when neural networks' learning flexibility and HSA's optimization capabilities are combined. Findings indicate that models with near-perfect accuracy, including Decision Tree Classifier and Sequential models, have the potential to transform fraud prevention.

2025, International Journal of Digital Innovation, Insight, and Information

Massive datasets and dynamic fraud patterns pose serious obstacles to financial transaction fraud detection. In order to provide adaptive fraud detection, this study suggests a hybrid model that combines Test Case Reduction, Gaussian Mixture Models (GMM), and Regularized Discriminant Analysis (RDA). GMM detects anomalies by capturing the probabilistic distribution of financial behaviors, whereas RDA improves classification accuracy, particularly with unbalanced datasets. Test Case Reduction reduces superfluous data, which increases efficiency. The accuracy, precision, and scalability of the model are improved over conventional methods. It is perfect for financial institutions looking to increase security because it instantly adjusts to new fraud practices. When compared to current models, experimental results show a significant improvement in performance indicators like accuracy (98.4%) and F1-score (97.0%). Furthermore, the hybrid model's longterm efficacy is guaranteed by its capacity to continuously adjust to changing fraud tactics. It offers a scalable method appropriate for large-scale financial transactions by utilizing probabilistic modeling and classification approaches powered by machine learning. To improve fraud detection accuracy and operational efficiency even more, future developments might incorporate reinforcement learning techniques and deep learning-based improvements.

2025, International Journal of Computer Science Engineering Techniques

Fraudulent financial transactions pose a critical challenge in the banking sector, necessitating robust and adaptive fraud detection mechanisms. This study proposes a Transformer-Based Sequential Fraud Detection model that efficiently detects fraudulent activities by capturing long-range dependencies in financial transactions. Utilizing the Synthetic Financial Datasets for Fraud Detection (PaySim), the model is trained to identify anomalies through sequential analysis of transaction patterns. Performance evaluation demonstrates 99.20% accuracy, 99.02% precision, 99.41% recall, and an F1-score of 99.21%, outperforming traditional fraud detection methods. The model achieves an AUC-ROC of 0.9928 and a Precision-Recall AUC of 0.9909, confirming its effectiveness in minimizing false alarms. A detailed analysis of the confusion matrix further highlights its real-world applicability in reducing financial fraud risks. These findings establish the proposed model as a highly efficient and scalable approach for fraud detection in digital banking.

2025

This research has an information technology background that plays an important role in human life both in the present and in the future. The internet can penetrate the boundaries between countries and accelerate the spread and exchange of knowledge both among scientists or scholars throughout the world. Behind the ease of use of the internet, there is a dark side that worries the users, namely in terms of security. Public and global internet networks are very vulnerable to various forms of crime. The actions taken by the government to overcome in terms of reducing crime activities enter or infiltrate into a computer network illegally is the issuance of Law NO.11 of 2008 concerning Information and Information Transactions can be interpreted as a crime committed by a person or group of people with the intention of taking the advantage of others through a computer network. This study uses legal normative juridical research. The purpose of this study was to find out how the modus operan...

2025

2025, International Journal of Accounting and Economics Studies

The rise of digital payments has accelerated the need for intelligent and scalable systems to detect fraud. This research presents an end-toend, feature-rich machine learning framework for detecting credit card transaction anomalies and fraud using real-world data. The study begins by merging transactional, cardholder, merchant, and merchant category datasets from a relational database to create a unified analytical view. Through the feature engineering process, we extract behavioural signals such as average spending, deviation from historical patterns, transaction timing irregularities, and category frequency metrics. These features are enriched with temporal markers such as hour, day of week, and weekend indicators to expose all latent patterns that indicate fraudulent behaviours. Exploratory data analysis (EDA) reveals contextual transaction trends across all the dataset features. Using the transactional data, we train and evaluate a range of unsupervised models: Isolation Forest, One Class SVM, and a deep autoencoder trained to reconstruct normal behavior. These models flag the top 1% of reconstruction errors as outliers. PCA visualizations illustrate each model's ability to separate anomalies into a two-dimensional latent space. We further segment the transaction landscape using K-Means clustering and DBSCAN to identify dense clusters of normal activity and isolate sparse, suspicious regions. Finally, we propose a composite risk score by aggregating binary flags from all anomaly detectors, unexpected spend indicators, rapid-use events, and high-frequency "spending sprees". This score highlights the riskiest cardholders and merchants, enabling prioritized investigation. Our framework detects approximately 1-2% of transactions as anomalies and effectively surfaces high-risk entities, demonstrating the power of unsupervised analytics for real-time fraud surveillance in dynamic financial ecosystems.

2025, Social Sciences

This article presents findings from a four-year collaborative research project on immigrant and mixed-status families in Santa Cruz County, California. The project employed a new model of critical community-engaged scholarship called Community Initiated Student Engaged Research (CISER) in order to gain access to and build trust with this vulnerable population. The study used an overarching theoretical framework of "belonging" to identify six key factors most consequential for belonging and/or exclusion, including access to education, economic security, legal immigration status, health services, opportunities for youth, and social networks. The findings reveal the complex and interconnected nature of these factors and demonstrate how exclusion experienced due to a lack of legal immigration status had far-reaching effects on interviewees' job prospects and experiences of economic, health, and housing insecurity. The article highlights the importance of using an assetsbased approach to draw out the myriad ways interviewees and communities create spaces, networks, and ways to promote and enhance both material and emotional forms of belonging. The CISER model and its participatory approach also provide tangible benefits for community partners and undergraduate researchers. This article contributes to the literature on immigrant experiences and critical community-engaged research while offering insights into sources of and systemic barriers to collective belonging.

2025, Journal of Consumer Research

Some food items that are commonly considered unhealthy also tend to elicit impulsive responses. The pain of paying in cash can curb impulsive urges to purchase such unhealthy food products. Credit card payments, in contrast, are relatively painless and weaken impulse control. Consequently, consumers are more likely to buy unhealthy food products when they pay by credit card than when they pay in cash. Results from four studies support these hypotheses. Analysis of actual shopping behavior of 1,000 households over a period of 6 months revealed that shopping baskets have a larger proportion of food items rated as impulsive and unhealthy when shoppers use credit or debit cards to pay for the purchases (study 1). Follow-up experiments (studies 2-4) show that the vice-regulation effect of cash payments is mediated by pain of payment and moderated by chronic sensitivity to pain of payment. Implications for consumer welfare and theories of impulsive consumption are discussed. T he past two decades have witnessed a rapid increase in obesity among U.S. consumers. According to the Center for Disease Control, 34% of U.S. adults are obese, up from 23% in 1988. An additional 33% are overweight . These results suggest that the consumption of unhealthy food is increasing and have prompted

2025

This paper introduces the major components of a SIP-based VoIP platform, referred to as NTP VoIP platform. Based on the NTP VoIP platform, the researchers can develop and deploy their applications and services. For lawful interception, we propose a monitoring system that provides call detail records and interception function. We also propose a conference system for audio and video conferences. In this paper, we provide detailed message flows to show how the monitoring system and the conference system work.

2025, International Journal of Computer Applications

Loan default prediction is one of the most important and critical problem faced by many banks and other financial institutions as it has a huge effect on their survival and profit. Many traditional methods exist for mining information about a loan application and have been greatly studied and applied in the past. These methods seem to be underperforming as there have been reported increases in the amount of bad loans and defaulters among many financial institutions. In this paper, gradient boosting algorithm called XGBoost was used for loan default prediction. The prediction is based on a loan data from a leading bank taking into consideration data sets from both the loan application and the demographic of the applicant. Similarly, important evaluation metrics such as Accuracy, Recall, precision, F1-Score and ROC area of the analysis were used. The paper provides an effective basis for loan credit approval in order to identify risky customers from a large number of loan applications using predictive modeling. The full utilization of this model will assist financial institutions in knowing a risking customer that may default in loan payment before lending.

2025, Information Fusion

We propose a novel approach for credit card fraud detection, which combines evidences from current as well as past behavior. The fraud detection system (FDS) consists of four components, namely, rule-based filter, Dempster-Shafer adder, transaction history database and Bayesian learner. In the rule-based component, we determine the suspicion level of each incoming transaction based on the extent of its deviation from good pattern. Dempster-Shafer's theory is used to combine multiple such evidences and an initial belief is computed. The transaction is classified as normal, abnormal or suspicious depending on this initial belief. Once a transaction is found to be suspicious, belief is further strengthened or weakened according to its similarity with fraudulent or genuine transaction history using Bayesian learning. Extensive simulation with stochastic models shows that fusion of different evidences has a very high positive impact on the performance of a credit card fraud detection system as compared to other methods.

2025, JPS (Jurnal Perbankan Syariah)

Pengaruh literasi keuangan syariah terhadap perencanaan keuangan syariah mahasiswa universitas Islam Abstrak Tujuan-Penelitian ini bertujuan menganalisa pengaruh literasi keuangan syariah mahasiswa universitas Islam terhadap perencanaan keuangan syariahnya. Metode-Penelitian ini menggunakan pendekatan kuantitatif dengan metode pengambilan sampel convenience sampling dan teknik analisis Structural Equation Model-Partial Least Square (SEM-PLS). Penelitian ini menganalisis 110 set data primer menggunakan aplikasi Smart PLS. Mewakili populasi mahasiswa universitas Islam, sampel atau responden adalah para mahasiswa dari program studi perbankan syariah di Universitas Islam Negeri Syarif Hidayatullah Jakarta sebagai salah satu universitas Islam negeri terbesar di Indonesia. Temuan-Hasilnya menunjukkan bahwa pengetahuan dan perilaku keuangan syariah mahasiswa berpengaruh positif terhadap perencanaan keuangan syariah, sedangkan sikap keuangan syariah mahasiswa tidak berpengaruh. Implikasi-Secara teoritis, penelitian ini dapat melengkapi teori yang sudah ada terutama pengaruh pengetahuan dan perilaku keuangan syariah terhadap perencanaan keuangan syariah. Secara praktis, penelitian ini dapat menjadi referensi dalm meningkatkan literasi dan perencanaan keuangan syariah mahasiswa universitas Islam, khususnya dari segi pengetahuan dan perilaku keuangan syariah.

2025, Marmara Üniversitesi Hukuk Fakültesi Hukuk Araştırmaları dergisi

Son yıllarda bankaların, kredi kartı kullanıcısı olan tüketicilerin hesaplarından "kart üyelik ücreti" adı altında kestikleri yıllık üyelik aidatının iptaline karşı açılan davalar, TKHK m. 6'da öngörülen sözleşmedeki haksız şartlar" kavramını ve bu kavramla bağlantılı olan diğer hukuki sorunları bir kez daha gündeme getirmiştir. Bu çalışmada, Yargıtay'ın 02.05.2008 tarihli kararına konu olan somut olaydan hareketle, tüketici sözleşmelerindeki haksız şartların hakim tarafından tespiti, haksız şartların denetlenmesinde öngörülen kriterler, sözkonusu haksız sözleşme koşullarının hükümsüzlüğü, bu hükümsüzlüğün türü ve bu sonucun sözleşmenin bütününe olan etkisi irdelenmeye çalışılacaktır. Bu çerçevede ayrıca, aynı hukuki sorunla karşı karşıya kalan diğer tüketicilerin etkin korunması açısından mevcut yasal düzenlemeler ve bu duruma ilişkin çözüm önerileri de ele alınacaktır.

2025

The rapid evolution of smart card technologies has intensified the need for efficient, portable solutions to access, extract, and analyze embedded data. This study presents the design and development of a Portable Smart Card Reader equipped with automated data extraction and analysis capabilities, aiming to enhance real-time data processing and mobility across various application domains such as healthcare, finance, and security. The proposed system integrates a compact hardware module with embedded microcontroller architecture, contact/contactless card compatibility, and a user-friendly interface. Advanced software algorithms were developed to enable secure authentication, rapid data retrieval, and intelligent analysis of smart card content without requiring manual intervention. The prototype supports interoperability with standard smart card protocols (ISO/IEC 7816 and ISO/IEC 14443) and leverages cloud-based data synchronization for extended functionality. Performance evaluation revealed high accuracy, speed, and reliability in diverse operational environments. This innovation offers significant potential for institutions seeking scalable, cost-effective, and mobile smart card solutions with enhanced data intelligence capabilities.

2025

Background: Recent achievements in nano technology have led to the development of the target system of drug delivery. However, targeting a molecule to a particular Site using a drug delivery system effectively requires a specialized drug delivery system. The discovery of Nano sponge has become a significant step in overcoming certain problems such as drug toxicity, poor bioavailability and release of drug in a predictable fashion as they can accommodate both hydrophilic and hydrophobic drug. Methods : Nano sponge's technology has been used widely for the delivery of drugs for, topical Administration. In order to incorporate the material of choice between the Nano sponges, a solvent-mediated dissemination or interfacial solvent evaporation approach is used once the plant extract has been prepared. Usually made from Vinca rosea leaves, the extract aids in the formation of the nanosponges. The therapeutic use of nanoemulsion platforms for topical administration is limited by these drawbacks. A method for resolving this issue is the addition of nanoemulsion to a gelling solution. Nano sponges can also serve as an effective carrier for enzyme, proteins, vaccine and antibodies. The nano sponge is essentially a porous structure, which has a unique ability to capture pieces of medicine and ensures the dignity of liberation of desire. Nano sponges are tiny sponges that can circulate in the body to reach the specific site and binds on the surface to release the drug in a controlled and predictable manner.

2025, Retrieved September

2025, THE RESEARCH JOURNAL (TRJ)

In today's data-driven environment, the security of databases has become a critical concern for organizations across all sectors. With increasing incidents of cyber-attacks, data breaches, and insider threats, there is an urgent need for systematic and proactive approaches to safeguard sensitive information. Database security audits serve as a vital mechanism to identify, assess, and mitigate vulnerabilities before they can be exploited by malicious actors. This paper presents an in-depth study on the role and methodology of security audits in enhancing database security, with a special focus on vulnerability detection and prevention. The audit process involves a structured assessment of database configurations, access controls, user privileges, logging mechanisms, and patch management strategies. By employing advanced auditing tools and techniques, organizations can uncover hidden threats, non-compliant practices, and configuration weaknesses that could potentially lead to security breaches. Moreover, periodic audits enable compliance with regulatory standards such as GDPR, HIPAA, and PCI DSS, ensuring both legal and operational security. The paper includes a comprehensive literature survey that explores existing research, tools, and frameworks used in database security audits. It also discusses core working principles, including automated scanning, anomaly detection, and real-time monitoring. Case studies and comparative analyses demonstrate the effectiveness of various audit practices in both Oracle and open-source database systems. The study emphasizes the need for continuous enhancement in audit methodologies through artificial intelligence and machine learning. These emerging technologies have the potential to revolutionize audit efficiency and threat prediction. The findings underscore the importance of incorporating proactive security audits as an integral component of database security strategy to prevent breaches before they occur.

2025, International Journal of Management, Technology And Engineering

In the modern digital landscape, data has become one of the most critical assets for organizations, making databases a primary target for increasingly sophisticated cyber threats. Oracle Database, being one of the most widely used relational database management systems (RDBMS), plays a crucial role in managing sensitive and mission-critical data across industries. Consequently, ensuring the security of Oracle databases is imperative for protecting data integrity, confidentiality, and availability.This paper explores advanced Oracle security techniques designed to combat evolving cyber threats that target database infrastructures. Traditional security mechanisms, while still relevant, are no longer sufficient in isolation due to the complexity and persistence of modern attacks. Oracle

2025, Science, Technology and Development

In today's data-driven world, the integrity, confidentiality, and availability of enterprise data are critical. Oracle databases, widely adopted by large-scale organizations across various sectors, play a fundamental role in managing and storing this sensitive information. However, as the reliance on Oracle systems grows, so does the surface area for potential security threats. This paper explores

2025, Science, Technology and Development

In the digital era, where information is a critical business asset, ensuring the security of enterprise databases has become a paramount concern. Among the various database management systems in use today, Oracle Database stands out for its comprehensive, multi-layered security architecture designed to protect data integrity, confidentiality, and availability. This paper, "Mastering Oracle Database Security: Best Practices for Enterprise Protection," delves into the critical security mechanisms within Oracle's ecosystem and offers practical guidance on implementing them effectively in enterprise environments.The focus of this research is to explore the core principles and best practices for securing Oracle databases against internal and external threats. As data breaches and cyberattacks become more advanced and frequent, organizations must adopt robust security measures that go beyond conventional perimeter defenses. Oracle provides a wide range of tools and featuresincluding Transparent Data Encryption (TDE), Oracle Label Security (OLS), Virtual Private Database (VPD), Database Vault, auditing mechanisms, and secure backup and recovery procedures-to help enterprises safeguard sensitive data.This paper systematically examines the foundational components of Oracle's security architecture, including authentication and authorization models, Role-Based Access Control (RBAC), fine-grained access control, and encryption of data at rest and in transit. The study also evaluates the role of auditing tools, such as Fine-Grained Auditing (FGA) and Unified Audit Trail, in ensuring accountability and compliance with regulatory standards like GDPR, HIPAA, and PCI DSS.A literature survey provides context through the evolution of database security practices and compares Oracle's offerings with those of other major database platforms. Furthermore, realworld examples and breach analyses highlight the importance of proper configuration, continuous monitoring, and policy-driven governance in minimizing risk.Finally, the paper explores emerging trends and future enhancements, such as the integration of machine learning for adaptive threat detection, Zero Trust security models, and enhanced security in Oracle Autonomous Databases. The insights presented aim to help organizations make informed decisions about their data security strategies and reinforce Oracle's role as a leader in enterprise database protection.

2025, International Journal for Research in Applied Science and Engineering Technology

2025, International Journal of Software Engineering and its Applications

This paper discussed the past works on fraud detection system and highlights their deficiencies. A probabilistic based model was proposed to serve as a basis for mathematical derivation for adaptive threshold algorithm for detecting anomaly transactions. The model was optimized with Baum-Welsh and hybrid posterior-Viterbi algorithms. A credit card transactional data was simulated, trained and predicted for fraud. And finally, the proposed model was evaluated with different metric. The results showed that with the optimization of parameters, posterior-Viterbi cum new detection model performed better than Viterbi cum old detection model.

2025

Access to credit plays a vital role in the growth and sustainability of fish farming operations. It enables farmers to invest in essential inputs such as quality seeds, feed, equipment, and infrastructure. Adequate financing supports improved operational efficiencies, enhances productivity, and helps farmers to scale their businesses. Rank Based Quotients (RBQ) was used to study the constraints faced by fish farmers of Kishanganj district. Creditworthiness and Documentation, Lack of financial literacy, Uncooperative behavior of Bankers and complex application procedure for credit, were identified as constraints faced by fish farmers of Kishanganj district to avail the institutional credit facilities.

2025, Wei Dai

Bitcoin whitepaper'ında ilk referans olarak gösterilen dokümanın resmi olmayan Türkçe çevrilmiş halidir. Wei Dai tarafından kaleme alınmış olup Yunus Emre Özbucak tarafından çevrilmiştir.

2025, International Journal of Computer Science and Mobile Computing

In the era of increasing cyber threats, the implementation of robust Intrusion Detection Systems (IDS) is crucial for safeguarding network integrity. This study presents a comprehensive comparison of various machine learning algorithms employed in IDS, with a specific focus on binary logistic regression as a comparative tool. We utilized a well-established dataset to evaluate the performance of multiple algorithms, including decision trees, support vector machines, and neural networks. Our findings reveal significant variations in accuracy, precision, and recall across the different algorithms. Binary logistic regression served as an effective benchmark, highlighting the strengths and weaknesses of each model. This research contributes to the ongoing discourse in cybersecurity by providing empirical evidence on the efficacy of machine learning approaches in detecting intrusions, offering insights for future enhancements in IDS design.

2025, Journal of Financial Counseling and Planning

The shift from defined benefit to defined contribution retirement plans, increasingly complex and rapidly changing tax laws, greater household personal wealth within cohorts, along with a broad array of financial products available to transfer resources across the life cycle, have been both a burden and an opportunity for many households . The burden is that a failure to understand the financial planning process (i.e., determining financial goals, managing and protecting resources, etc.) may lead to significant welfare loss. Increasing choice and complexity also can also benefit those with sufficient in-

2025, International Journal for Research in Applied Science & Engineering Technology (IJRASET)

Traditional attendance systems are vulnerable to fraudulent practices and inefficiencies. This paper proposes a facial recognition-based attendance tracking system enhanced with anti-spoofing capabilities using YOLOv5. By integrating real-time face detection, recognition, and liveness verification, the system ensures accurate and secure attendance logging. It further integrates with a web portal via ThingSpeak API to notify absenteeism.

2025

Financial technology (fintech) applications increasingly rely on continuous data streams from Internet-of-Things (IoT) devices-such as payment terminals, mobile banking platforms, and point-of-sale (POS) systemsto drive real-time analytics. This paper addresses the challenges of real-time data analytics in fintech, including strict low-latency requirements, high-volume heterogeneous data, and stringent security and compliance demands. We propose an edge-cloud collaborative architecture that distributes analytic workloads between edge devices (near data sources) and the cloud, to enable timely processing of IoT data streams without sacrificing scalability or accuracy. The proposed architecture is tailored to fintech use cases, with an emphasis on instant fraud detection, transaction monitoring, and customer experience optimization. We design and evaluate the architecture through a prototype implementation, including data flow diagrams, a layered processing pipeline, and a performance evaluation measuring latency, throughput, and scalability. Experimental results demonstrate that the edge-cloud approach significantly reduces end-to-end latency (often by an order of magnitude) and improves throughput under load, compared to a cloud-only deployment. We also discuss how the architecture supports continuous model training and adaptation, data security (keeping sensitive data at the edge when possible), and regulatory compliance. Relevant methods and technologies co-authored by Akash Vijayrao Chaudhari-including IoT data warehousing, federated learning for distributed analytics, and AI-driven fintech anomaly detection-are integrated and cited to situate our contributions in the state-of-the-art. The paper concludes that an edge-cloud collaborative paradigm is a promising foundation for next-generation fintech analytics systems, combining the agility of edge computing with the power of cloud-scale data processing.

2025, IEEE

Online banking, credit card purchases, and mobile wallet are all examples of the growing popularity of banking activities. Financial organisations, retailers, and consumers all face the formidable problem of detecting fraudulent financial transactions. The traditional rulebased detection methods are frequently insufficient due to the growing sophistication of fraudulent activities. The development of AI has made it possible to intelligently analyse massive amounts of financial data using machine-learningbased algorithms in order to spot fraudulent activities. Recent studies have shown that ML approaches may effectively identify fraudulent transactions in massive volumes of payment data, which is a crucial component of cyber-crime organisations' operations. This research provides machine learning models to detect fraudulent financial transactions using the PaySim dataset. Label encoding, min-max scaling, and SMOTEbased balancing are utilized to overcome the intrinsic class imbalance in fraud detection tasks. Classification models like SVM, NB, RF, and LR are assessed using F1-score, recall, accuracy, and precision. The outcomes indicate that Logistic Regression excels in accuracy with 98.99%, whereas Random Forest excels in recall and Naïve Bayes excels in precision. The study emphasizes the need for strong fraud detection systems and the trade-offs involved in optimizing evaluation measures to combat evolving financial crimes.

2025, Journal of Business Research

This article examines how men and women differ in both their perceptions of the risks associated with shopping online and the effect of receiving a site recommendation from a friend. The first study examines how gender affects the perceptions of the probability of negative outcomes and the severity of such negative outcomes should they occur for five risks associated with buying online (i.e., credit card misuse, fraudulent sites, loss of privacy, shipping problems, and product failure). The second study examines gender differences in the effect of receiving a recommendation from a friend on perceptions of online purchase risk. The third study experimentally tests whether, compared to men, women will be more likely to increase their willingness to purchase online if they receive a site recommendation from a friend. The results suggest that, even when controlling for differences in Internet usage, women perceive a higher level of risk in online purchasing than do men. In addition, having a site recommended by a friend leads to both a greater reduction in perceived risk and a stronger increase in willingness to buy online among women than among men.

2025, XXVI Congreso Argentino de Ciencias de la Computación (CACIC) (Modalidad virtual, 5 al 9 de octubre de 2020)

En el estudio de los algoritmos de Minería de Datos del tipo supervisados surge el problema del desbalance de clases, que implica que la información no se encuentre distribuida equitativamente entre todas las clases que la componen, por lo que se generan efectos no deseados en el proceso de clasificación. Este trabajo considera el caso de conjuntos de datos que solamente tiene dos clases y una de ellas cuenta con una mayor cantidad de ejemplos que la otra. El interés principal del trabajo es la aplicación de la técnica de balanceo de clases SMOTE (Synthetic Minority Oversampling Technique), que con algoritmos de interpolación incrementa en forma "sintética" los ejemplos de la clase minoritaria. Los resultados experimentales muestran que algunas técnicas, en el proceso de entrenamiento, obtienen mejores porcentajes de clasificación, cuando se usan estos datos artificiales. El

2025, International Congress Series

In order to produce database-eligible DNA profiles from touched objects, each testing procedure including sample recovery, extraction, amplification and separation was evaluated and optimized. The developed methodologies were tested on control samples as well on fingerprints deposited on a variety of substrates such as credit cards, keys, and pens. All samples were amplified in triplicate to confirm the presence of each allele and to detect drop-ins. Overall the modifications implemented produced reproducible results for DNA titrated to 20 pg. For DNA dilutions, 25 pg routinely resulted in full profiles, and 12.5 pg determined 76.9% of the database loci tested. Similarly, for the touched objects, 75.8% of the 20-pg to 100-pg samples yielded database-eligible profiles; the remaining samples either were mixtures or contained an insufficient number of allelic calls. Here, the three-amplification approach was crucial and produced more complete profiles with confidence in the allelic assignments. DNA amounts below 20 pg did show partial profiles with correct allelic determinations that could have been compared in a specific case but were often too incomplete for database entry.

2025

In this paper we examine the problem of infer ence in Bayesian Networks with discrete random variables that have very large or even unbounded domains. For example, in a domain where we are trying to identify a person, we may have vari ables that have as domains, the set of all names, the set of all postal codes, or the set of all credit card numbers. We cannot just have big tables of the conditional probabilities, but need compact representations. We provide an inference algo rithm, based on variable elimination, for belief networks containing both large domain and nor mal discrete random variables. We use inten sional (i.e., in terms of procedures) and exten sional (in terms of listing the elements) represen tations of conditional probabilities and of the in termediate factors.

2025, The Journal of Finance

Qualitative choice models of consumers' decisions to file for bankruptcy and their choice of bankruptcy chapter are estimated jointly, combining choice‐based sampling techniques with a nested estimation procedure. Medical and credit card debt are found to be the strongest contributors to bankruptcy, with homeownership playing an important role with respect to both the decision to declare bankruptcy and the choice of bankruptcy alternative. The potential effects of legal changes relating to property exemptions and dischargeable debt categories are found to encourage debt repayment through Chapter 13.

2025, International Journal of Design

Involving users in the design process is increasingly discussed as the quickest and most reliable way to capture the needs of users and consumers. In parallel, the fastest growing population segment in Asia and the West is older people. This article asks whether their involvement in the design process could accelerate a growing service market and if so, how? It addresses a knowledge gap that constrains service provision for a growing market of older people and which underestimates older people's potential contribution in the early phases of the development of new services. The current role of older users is limited to that of test persons later in the design process or as objects of randomized samples that explore consumers' reactions to existing products. The present case study provides an empirical example of how old users can be involved in the early stages of service design. In doing this, the article questions the concept of old users as laggards. It suggests great potential to include such users -been arounds -as sources of innovation in the earlier phases of the design process if they have the right tools and opportunities to act. In identifying unsatisfied needs and potential market solutions, the inclusion of old users in user-driven projects can contribute to the generation of business ideas.

2025

This paper introduces a novel conceptual model for quantum information processing based on an optical framework involving controlled polarization of photon beams projected onto a spatial pixel array. The model reinterprets Qubit evolution and interference through a visual analogy: a dynamic film of controlled quantum states interacting with an encoded image. We explore both sequential (scanning) and parallel (full-array) projection architectures, analyze their implications for entanglement and interference, and show how they may simulate pattern recognition via quantum processes. The proposed model provides a physically intuitive alternative to conventional gate-based computation, with potential for pedagogical use and architectural inspiration.

2025, International Research Journal Of Modernization in Engineering Technology and Science

Chemotherapy-induced myelosuppression (CIM) is a major concern in pediatric oncology, often leading to severe complications, including infections, anemia, and thrombocytopenia. Predicting CIM accurately can help in optimizing treatment plans and minimizing adverse effects. This research explores the development of a Generative AI model that leverages deep learning to predict and categorize CIM in children undergoing chemotherapy. The model integrates clinical data, blood count trends, and patient history using a Transformerbased generative framework. Our indings suggest that the AI model signi icantly improves prediction accuracy and provides actionable insights for personalized treatment adjustments.

2025, International Research Journal of Modernization in Engineering Technology and Science

Substance abuse remains a critical public health challenge in the United States, impacting individuals, families, and communities. Behavioral therapies, a cornerstone of substance abuse treatment, have seen advancements with the integration of technology. This paper delves into the application of Generative Artificial Intelligence (Gen AI) in enhancing behavioral therapies for substance abuse treatment and recovery. Gen AI, with its ability to generate personalized, adaptive, and context-aware responses, offers new avenues for addressing the unique needs of individuals in treatment programs.
The study outlines the architecture and mechanics of Gen AI algorithms, detailing their design, training methodologies, and real-world applications in therapeutic settings. It examines the integration of Gen AI in therapy delivery, including personalized virtual therapy sessions, real-time relapse prevention tools, and recovery support systems. Additionally, the paper provides step-by-step implementation guides, supported by practical code examples, flow diagrams, and illustrative figures, to demonstrate how these technologies can be deployed in treatment settings.
Furthermore, the paper explores the ethical considerations, challenges, and potential limitations of using Gen AI in substance abuse treatment, such as data privacy, bias mitigation, and the need for human oversight. By combining theoretical insights with practical applications, this study aims to inform researchers, clinicians, and policymakers on leveraging Gen AI to improve treatment outcomes, enhance recovery experiences, and ultimately address the ongoing substance abuse crisis in the United States.

2025, International Research Journal of Modernization in Engineering Technology and Science

This paper presents a substantially reworked examination of how advanced game-theoretic paradigms can serve as a foundation for the next-generation challenges in Artificial Intelligence (AI), forecasted to arrive in or around 2025. Our focus extends far beyond traditional zero-sum or Nash equilibrium models by incorporating novel dimensions such as dynamic coalition formation, language-based utilities, sabotage risks, and partial observability. In doing so, we provide a diverse set of mathematical formalisms, experimental simulations, and practical coding schemes that detail how multi-agent AI systems may evolve, adapt, and negotiate when confronted with complex real-world dilemmas. The paper includes extensive discussions on repeated games, Bayesian updates for adversarial detection, and the integration of moral or normative frames into classical payoff structures. We also present newly developed algorithms and proof sketches that clarify the convergence of multi-agent reinforcement learning toward game-theoretic equilibria in large, high-dimensional action spaces. By proposing both conceptual expansions and practical coding illustrations, we aim to equip AI researchers, engineers, and strategists with robust theoretical instruments for shaping the interplay between AI agents and human stakeholders in uncertain, partially adversarial contexts. This reworked discussion not only emphasizes the state of the art in game-theoretic AI but also indicates clear directions for future investigation.