Transaction Data Research Articles

EXPRESS: Buy Now Pay Later: Impact of Installment Payments on Customer Purchases

Buy Now Pay Later (BNPL) installment payments allow customers to pay for purchases in a series of interest-free installments over a short period of time. This research provides novel insights into how customer adoption of BNPL installment payments impacts spending. The authors leverage customer-level transaction data before and after the introduction of a BNPL installment payment service at a large US retailer. A difference-in-difference analysis indicates that the adoption of BNPL installment payments is associated with (1) an increase in purchase incidence, and (2) larger purchase amounts. These effects are statistically and economically significant over time. Moreover, this increase in spending is greater for smaller (vs. larger) basket shoppers, and for shoppers who relied more heavily on credit (vs. debit) cards before adoption. Three pre-registered experiments show that BNPL installment (vs. lump sum) payments increase spending by reducing perceived financial constraints. Specifically, BNPL installments alleviate perceived financial constraints by reducing perceived costs and facilitating budget control. These findings highlight the substantive role of BNPL installment payments in shaping purchase behavior and provide important implications for the management of BNPL payment schemes.

Leveraging tabular GANs for malicious address classification in ethereum network

The popularity of ethereum for cryptocurrency transactions attracts malicious actors to engage in illegal activities like phishing, ponzi, and gambling. Previous studies have focused mainly on phishing due to the large number of phishing addresses. However, there is no work done on ponzi or gambling classification due to the limited availability of these addresses, which makes their classification more challenging. In this paper, we propose a machine learning (ML) based method for classifying malicious addresses in ethereum, with a specific focus on phishing, ponzi, and gambling addresses. We use a selective upsampling technique through the tabular generative adversarial network (GAN) to solve limited data problems. We perform not only binary but also multiclass classification on various feature extraction methods, including Trans2Vec and Node2Vec, using Ethereum transactional data. We evaluate our method on F1 score, precision, recall, and accuracy. Our results show that the proposed method is effective in ponzi and gambling detection when compared with the state-of-the-art.

Clarifying to whom, where, and when payments are made : Recommendations from the ERPB Working Group on Transparency for Retail Payment End Users

It has become increasingly difficult for consumers to identify to whom, where, and when they have made payment transactions, as the descriptions provided on account statements do not always match the expectations of the consumer. Given that multiple actors in the payment chain can influence the information provided to the consumer, this problem can only be addressed with the involvement of multiple stakeholders. This paper discusses the recommendations made by the working group established by the Euro Retail Payments Board (ERPB) to address this issue. As regards the name of the party to whom payment has been made, the working group recommended that the commercial trade name of the payee (defined as the name used by the merchant to identify itself to the consumer) should be used along the entire payments chain. Regarding the where element, the working group recommended that the account statement should indicate the precise geographical location of the transaction. In the case of online commerce, the commercial trade name displayed on the website, or the commercial trade name of the payee’s online platform should be mentioned. Related to the when element, the working group recommended that the account statement should indicate clearly the date and time of the transaction as it would be known by the consumer. The working group called upon all actors involved to create and maintain quality transaction data and to ensure that the underlying technical standards being used are capable of supporting the collection, preservation and safe transmission of the data required by the consumer. The industry was expected to implement these voluntary recommendations by June 2024, but although progress has been made, the overall progress seems to be lagging slightly as the various stakeholders are addressing the issue in isolation. In the meantime, the European Commission has adopted some of the working group’s recommendations into proposed new payments legislation.

Credit Card Fraud Detection Using Machine Learning Algorithms

This study aims to address the rising issue of credit card fraud by developing a machine learning model capable of identifying and preventing fraudulent transactions. The model works by analyzing transaction data to detect potential fraud, subsequently canceling the transaction and alerting the credit card owner. Credit card fraud detection is a classification problem, where various machine learning algorithms are applied to distinguish between legitimate and fraudulent transactions. The analysis emphasizes the importance of robust countermeasures due to the increasing use of credit cards globally. However, real-world implementation of such systems may face challenges, particularly in securing the cooperation of banks and addressing resource constraints. The study also highlights key dataset features that correlate with fraudulent behavior, with ensemble methods standing out as top-performing algorithms in terms of accuracy and efficiency.

An Empirical Research on the Performance of Asset Pricing Models in the Chinese A-Share Market

Asset pricing remains a cornerstone of financial field, providing a framework to understand the anticipated returns of asset portfolios and individual assets. Its widespread application spans academic discourse and practical financial methodologies. This study conducts an empirical analysis of the Chinese A-share Market, utilizing transaction data from the A-share market spanning from 1994 to 2024. It examines the applicability of the traditional asset pricing models within the Chinese stock market. Through classic regression analyses of the CAPM, the Fama and French three-factor model (FF3), the Fama and French five-factor model (FF5), this study verifies that the FF5 provides a superior explanation of stock expectations for the A-share market during this period compared to other traditional asset pricing models. The conclusions of this study indicate that, among the CAPM, FF3, and FF5 models, the number of parameters taken into account by the model and its capacity to explain asset returns are positively correlated.

Competency or investment? The impact of NFT design features on product performance

This study investigates how NFT design features affect project performance. From the consumption perspective, NFT design features are divided into competency-related (image complexity, consistency) and investment-related (initial price, royalty). Using transaction data of 3297 NFT projects, we find that image complexity has an inverted U-shaped effect on long-term performance, while consistency boosts both short-term and long-term performance by formulating brand symbolism. Royalty as investment cost negatively affects short-term performance, while royalty and initial price exhibit inverted U-shaped impacts on long-term performance due to their mixed roles of costs and quality signals. Market uncertainty amplifies the impacts of complexity and royalties while diminishing the initial price impact in the long term. Findings support NFT project design, enhancing Web3 consumer behavior understandings.

Blockchain-Machine Learning Fusion for Enhanced Malicious Node Detection in Wireless Sensor Networks

In wireless sensor networks (WSNs), the presence of malicious nodes (MNs) poses significant challenges to data integrity, network stability, and system reliability. These issues are intensified by energy resource constraints and limitations within centralized authentication systems, necessitating an energy-efficient solution to ensure real-time responsiveness. Although artificial intelligence-driven approaches enhance detection capabilities, they overcome challenges related to data volume, coordination overhead, and latency in centralized control. This study introduces blockchain-machine learning (BC-ML), a novel hybrid model that seamlessly integrates blockchain and machine learning (ML) techniques to effectively identify MNs in WSNs. The model establishes an energy-efficient blockchain among cluster heads (CHs) for robust node authentication, incorporating a Schnorr-like zero-knowledge-proof technique to validate node data during communication initiation. Utilizing a hybrid lightweight approach with both symmetric and asymmetric ciphers enhances the security of node data transmission. A new proof-of-authority method is introduced, which leverages node digital certificates instead of conventional data transactions. This consensus mechanism reduces the processing overhead associated with larger data sizes in traditional proof-of-work methods, thereby improving both energy efficiency and scalability. To address dataset imbalances, the model employs a hybrid unsupervised ML technique, combining adaptive synthetic sampling with a convolutional neural network for efficient analysis of nodes and network features. The ML model, hosted on a robust data server, ensures ongoing oversight by updating CHs with security levels for detected MNs, thereby reducing storage and mitigating coordination challenges. Comprehensive analyses validate the effectiveness of the BC-ML model for detecting MNs, optimizing resource utilization, minimizing delays, and prolonging node and network lifetimes. Security analysis further confirms the ability of the model to mitigate diverse attacks and meet the stringent WSN security requirement.

Exploring the Intersection of E-commerce and Healthcare: A Visual Analysis of Research Trends.

The role of e-commercein healthcare is one of the key research areas emerging from the broader digital transformation of health services and consumer interactions. This study presents the bibliometric analysis of literature on e-commerce in healthcare based on publications from1993 to 2024 in Scopus. In particular, this analysis shows trends in the publication record that have been constantly growing, with prominent increases during the COVID-19 pandemic, now hardwiring the fact that digital technologies are key in promoting global health. The study surveys the thematic evolution of key technologies such as artificial intelligence (AI), which is transforming predictive healthcare analytics; blockchain, which is enhancing security in health data transactions; and telemedicine, which is revolutionizing remote patient care, alongside other emerging technologies that are reshaping global healthcare systems. The research also showcases global collaboration patterns, underlining the international dimensions of scholarly contributions to the area under review. These insights provide a foundation for understanding the current state of e-commerce in healthcare and suggest future directions for research and policy development in this fast-evolving field.

Recommendation System for Determining the Best Banner Supplier Using Profile Matching and TOPSIS Methods

Background: Choosing a banner supplier is a significant challenge for digital printing companies due to the various advantages offered by each supplier, often leading to selections based on subjective aspects such as price and quality. Objective: This research aims to develop a system that determines the best banner supplier to minimize production inefficiencies and maximize profits by comparing two calculation methods, Profile Matching and TOPSIS. Methods: A quantitative study was conducted using transaction data from the last six months. The parameter criteria used in this system include price, quality, delivery, availability, and payment terms. The study compares the effectiveness of Profile Matching and TOPSIS methods in identifying the best supplier. Results: The study results show that the TOPSIS method is superior, yielding 100% accuracy, 84% recall, and a 92% F1-score, outperforming the Profile Matching method. This demonstrates that the correct method and algorithm effectively provide the best alternative recommendations. Conclusion: The results indicate that using the TOPSIS method leads to more accurate and objective decisions based on predetermined criteria. The findings suggest that further research should focus on refining these methods to enhance decision-making in supplier selection.

Optimasi Penyusunan Koleksi Buku Dinas Perpustakaan Berdasarkan Pola Peminjaman dengan Metode Apriori

A library is a facility or place that provides reading materials. Good book arrangement can help the library in obtaining good reading sources. The arrangement of library service book collections based on borrowing patterns, there is an alignment between user needs and the availability of reading materials available in the library. Analysis of book borrowing patterns provides valuable insights for library staff in determining the books that are most in demand and often needed by users. Data mining is defined as mining data or efforts to dig up valuable and useful information in a very large database. The most important thing in data mining techniques is the rule for finding high frequency patterns between sets of itemsets called Association Rules. The method used in this study is Apriori (Association Rule). This technique is used to find relationships or associations between items or variables in data. Well-known algorithms such as Apriori and Eclat are used to find association rules in transactional data. The purpose of this study is to find out library visitor data using the Apriori Algorithm method and to find out the application of data mining for compiling book collections based on borrowing patterns. The results of this study are the multiplication of support and confidence, choose the one with the largest multiplication result. The largest result of the multiplication of these multiplications is the rule used when borrowing books. Because the results of the multiplication of the 4 borrowings have the same value, all of them can be used as rules.

Pengembangan sistem informasi penjualan berbasis web pada toko wallpaper dengan metode prototype

Sales information systems are a crucial component for companies, including Toko BIFFA Wallpaper, which requires increased efficiency in product sales. The current manual system causes various problems such as manual recording of transactions, income, and expenses, as well as ineffective archiving and management of daily and monthly financial reports. This research aims to build and implement a web-based sales information system to facilitate cashiers and owners of Toko BIFFA Wallpaper Kepanjen in managing, storing, and searching transaction data and reports. The method used is the prototype method, which includes needs analysis, rapid design, prototype construction, initial user evaluation, prototype improvement, and system development. The system is developed using PHP programming language and MySQL database, with testing conducted using the blackbox technique. The test results show that the system's appearance and features function well and meet user needs. This research concludes that the developed web-based sales information system significantly aids and facilitates cashiers and owners of Toko BIFFA Wallpaper Kepanjen in sales management, increasing efficiency and reducing manual errors.

Advancing credit risk assessment and financial decision-making: Integrating modern techniques and insights

Credit risk assessment and fraud detection are critical functions in the financial industry, necessary for ensuring the stability and integrity of financial institutions. Traditional approaches often struggle to accurately assess risk and detect fraudulent activities in a timely manner. However, the rise of machine learning has introduced powerful tools that leverage large datasets and advanced algorithms to improve these processes. This research paper investigates the application of machine learning techniques in credit risk assessment and fraud detection within financial transactions. The paper begins by emphasizing the importance of accurate risk assessment and fraud detection in the financial sector and introduces machine learning as a solution to the limitations of traditional methods. A thorough literature review examines existing methodologies, algorithms, and trends, highlighting the advancements in this field. The discussion covers data acquisition and preprocessing techniques, underscoring the necessity of clean and relevant data for effective model training. Additionally, feature engineering strategies are explored to extract valuable insights from financial transaction data, enhancing the predictive power of machine learning models. The research analyzes various machine learning algorithms, such as logistic regression, decision trees, random forests, support vector machines, and neural networks, as well as ensemble methods. Model evaluation metrics, including accuracy, precision, recall, and ROC-AUC, are employed to assess the performance of these algorithms. The paper presents case studies and experimental results to demonstrate the practical application of machine learning models in real-world scenarios, showcasing their effectiveness in improving credit risk assessment and fraud detection. Furthermore, the research explores challenges such as imbalanced datasets, model interpretability, and regulatory compliance, offering insights into potential solutions and future research directions. The integration of behavioral finance, Bayesian networks, and optimization methods is also discussed, highlighting how these modern approaches, combined with big data analytics, can enhance predictive accuracy and decision reliability. In conclusion, this research underscores the transformative potential of machine learning in credit risk assessment and fraud detection within financial transactions. By adopting advanced algorithms and data-driven approaches, financial institutions can significantly improve their risk management strategies, mitigate potential risks, and protect against fraudulent activities. This ultimately contributes to a more secure and resilient financial ecosystem, enabling institutions to maintain a competitive edge in an ever-evolving market.

Multi-attention recommender system for non-fungible tokens

Recommender systems have become essential tools for enhancing user experiences across various domains. While extensive research has been conducted on recommender systems for movies, music, and e-commerce, the rapidly growing and economically significant Non-Fungible Token (NFT) market remains underexplored. Recommender systems have the potential to significantly enhance user engagement, increase the time users spend on platforms, and deepen user involvement. Consequently, effective implementation of such systems could serve as a catalyst for invigorating the NFT market. However, the unique characteristics of the NFT market, such as the high sparsity of user–item interactions, anonymity of blockchain, and dual nature, present challenges not encountered in traditional recommender systems, highlighting the importance of developing tailored solutions to cater to its specific needs and unlock its full potential. In this paper, we examine the distinctive characteristics of NFTs and propose the first recommender system specifically designed to address NFT market challenges. In specific, we develop a Multi-Attention Recommender System for NFTs (NFT-MARS) with three key characteristics: (1) graph attention to handle sparse user–item interactions, (2) multi-modal attention to incorporate feature preference of users, and (3) multi-task learning to consider the dual nature of NFTs as both artwork and financial assets. We demonstrate the effectiveness of NFT-MARS compared to various baseline models using the actual transaction data of NFTs collected directly from the blockchain for four of the most popular NFT collections.

Exploration of the Size Effect on Transaction Data of Non-publicly Traded EU Companies

Exploration of the Size Effect on Transaction Data of Non-publicly Traded EU Companies

Product personalization and brand retailer performance: The critical role of brand retailer‐upstream supplier control

AbstractBrand retailers are increasingly offering product personalization to accommodate the requests of individual customers. While prior research has examined product personalization from a customer's or a manufacturer's perspective, this study focuses on product personalization from a brand retailer's perspective. Under product personalization, brand retailers that outsource production to upstream suppliers face two significant challenges: product returns from downstream customers, and personalization costs charged by upstream suppliers. Employing an analysis of transaction data from an online women's wedding dress firm, this study finds that as product requests increase (i.e., more product features must be modified), personalization costs increase, but the likelihood of product returns decreases. Additionally, as time requests increase (i.e., delivery is more urgent), the likelihood of product returns increases. Furthermore, relationship‐specific process control exacerbates (i) the effect of product requests on personalization costs, and (ii) the effect of time requests on the likelihood of product returns. Finally, relationship‐specific outcome control and transaction‐specific control alleviate the impact of time requests on the likelihood of product returns while exacerbating the effect of product requests on personalization costs. The findings suggest that retailers can realize greater benefits from product personalization by selecting control mechanisms tailored to personalization requests.

Penerapan Algoritma Apriori untuk Rekomendasi Asuransi pada Nasabah

Jasindo Insurance Company is one of the insurance companies that receives insurance coverage both directly and indirectly, with ownership of 1 share of dwiwarna series A owned by the Republic of Indonesia and 424,999 shares of Series B owned by PT Bahana Pembinaan Usaha Indonesia (Persero). PT Asuransi Jasa Indonesia has several products and choices in choosing which insurance is needed by customers in agriculture, health, education and many more. Due to the large amount of competition in the business world, it requires management to find the right strategy in increasing the use of Jasindo insurance by knowing the relationship between age, gender, marital status, occupation and the type of Jasindo insurance that is widely chosen by customers. In order to find out the use of insurance that is widely used by the community, it is necessary to analyze the data on the use of insurance using the apriori algorithm method to determine the combination between item-sets of transaction data on Jasindo insurance data. Based on the research conducted after experimenting with the above case with a minimum support = 25%, confidence = 100% so that the results of the rule that meets the support and confidence values are obtained, namely if the gender is male, the marital status is Unmarried, then the type of insurance is jasimdo health and jasindo rainbow then giving value is successful with 25% support, 100% confidence.

Manipulating stock prices based on insider information

Purpose This paper aims to investigate the trading behavior of insider investors before and after information releases, identifying information-based manipulation in the stock market and the characteristics of companies whose stock prices are manipulated. Design/methodology/approach This paper employs logit regression method and an event study approach, utilizing hand-collected data from 2010 to 2022, with information categorized into negative and positive types. Findings The results show no evidence of insider trading or negative information-based manipulation in both high and low transparency firms. However, in highly transparent companies, the Board of Directors (BOD) avoids direct manipulation by using relatives to evade market supervisors. In low transparency companies, both the BOD and family members (FM) exploit positive information to benefit personally by buying shares before releasing favorable news, causing a sharp stock increase, and selling afterward. Continued buying by the BOD and FM also suggests likely positive news announcements. Practical implications The characteristics of information-based manipulation in companies, as provided by this study, help individual investors avoid investing in stocks that are highly susceptible to manipulation. Originality/value Empirical research on information-based manipulation is scarce due to limited secondary data. Our study uses transaction data from insider investors in a frontier market with low transparency and high information asymmetry. This enables us to analyze information-based stock price manipulation. We identify manipulation by comparing insiders' trading behavior with their market information releases, resulting in stock price fluctuations greater than 5%.

Rancangan Sistem Informasi Penjualan Furniture pada Toko Mebel Pondok Baru Berbasis Web

Pondok Baru Furniture Store located in Makamhaji Kartasura is a business that operates in the field of manufacturing and selling furniture. There are problems in managing sales transaction data, stock of goods, and sales reports which still use manual methods, sometimes resulting in errors in recording. The current purchasing system which requires coming to the shop to see and buy the products being sold is very troublesome. And consumers want to know about the products being sold without having to come to the shop directly. Therefore, researchers created a web-based sales information system to overcome this problem. The method used in system development is the prototype method with data collection through interviews. The result is a web sales information system design that makes it easier for buyers to access products offered anywhere and carry out purchase transactions online. This system design is expected to produce more accurate data in processing sales data and sales reports.

Adopting Artificial Intelligence to Strengthen Legal Safeguards in Blockchain Smart Contracts: A Strategy to Mitigate Fraud and Enhance Digital Transaction Security

As blockchain technology increasingly underpins digital transactions, smart contracts have emerged as a pivotal tool for automating these transactions. While smart contracts offer efficiency and security, their automation introduces significant legal challenges. Detecting and preventing fraud is a primary concern. This paper proposes a novel application of artificial intelligence (AI) to address these challenges. We will develop a machine learning model, specifically a Convolutional Neural Network (CNN), to effectively detect and mitigate fraudulent activities within smart contracts. The AI model will analyze both textual and transactional data from smart contracts to identify patterns indicative of fraud. This approach not only enhances the security of digital transactions on blockchain platforms but also informs the development of legal standards and regulatory frameworks necessary for governing these technologies. By training on a dataset of authentic and fraudulent contract examples, the proposed AI model is expected to offer high predictive accuracy, thereby supporting legal practitioners and regulators in real-time monitoring and enforcement. The ultimate goal of this project is to contribute to legal scholarship by providing a robust technological tool that aids in preventing cybercrimes associated with smart contracts, thereby laying a foundation for future legal research and development at the intersection of law, technology, and security.

Identification and Analysis of Credit Card Fraud Based on Machine Learning Methods

Credit card has become an indispensable payment tool in People's Daily life. However, credit card fraud has also increased, resulting in huge financial losses for banks and individuals. Based on the real world credit card transaction data, this paper uses a variety of machine learning algorithms to identify and analyze credit card fraud. Firstly, the problem of missing and unbalanced data is solved through data preprocessing and feature selection, and important features are screened out, Such as transaction amount, product code, payment card type, etc. Then, integrated logistic regression, random forest, LightGBM, XGBoost, and Stacking models were constructed, and the models were trained and evaluated through cross-validation and parameter optimization. The experimental results show that the integrated Stacking model performs best in credit card fraud detection, with an AUC(Area Under the Curve) value of 0.960, which can accurately identify fraudulent transactions and provide an effective fraud warning mechanism for financial institutions. This mechanism significantly benefits both financial institutions and credit card users by reducing fraud losses and enhancing transaction security. However, it is acknowledged that the study has limitations, including the potential impact of data imbalance on model performance, the need for further testing of model generalization ability to new fraud patterns, and the importance of optimizing real-time performance to ensure practical applicability.