Transaction Patterns Research Articles

Machine learning and network analysis for financial crime detection: Mapping and identifying illicit transaction patterns in global black money transactions

The detection and prevention of malicious financial activities should be paramount for organizations in the US. Global economic integration, online banking, and increasing cases of cryptocurrency transactions have just increased the complexity of tracing illegal transactions. This research project examines the combined application and deployment of machine learning and network analysis in detecting black money transactions in the USA and globally. Machine learning and network analysis have emerged as a powerful mechanism in the fight against financial crime. Machine learning techniques, whereby systems learn through supervised and unsupervised learning, differ in that they can recognize patterns of financial data indicative of potentially fraudulent behavior. On the other hand, network analysis is one of the unique methods of detecting financial crimes, which derives power from presented relationships and interactions between sets of entities constituting transactional networks, such as people, companies, and accounts. This study used the Global Black Money Transactions dataset which revolved around financial transaction records involving unreported or illicit money, frequently for evading taxes, laundering money, or conducting illegal activities. Data come from financial institutions, government surveillance, whistleblowers, or investigations by the concerned law enforcement agencies. Rigorous data preprocessing steps were performed for the machine learning pipeline. In the current research project experiment, three machine learning algorithms were used: Logistic Regression, Random Forest, and XG-Boost. The performance indicators involved a set of standard metrics, including accuracy, precision, recall, F1 score, and AUC, which stands for Area Under the Curve. Despite lower accuracy, the XG-Boost algorithm was the best-performing algorithm. In terms of Precision, it correctly detected and predicted crimes, while the other models failed. Concerning the F1 Score, XG-Boost had the highest F1 score, balancing precision and recall. As per the AUC outcome, slightly better than Random Forest, XG-Boost was more capable of distinguishing between crime and non-crime transactions. Keywords: Black Money Transactions, Financial Crime, Machine Learning, Network Analysis, Mapping Illicit Transactions, XG-Boost, Random Forest, Logistic Regression

Intelligent Anti-Money Laundering Transaction Pattern Recognition System Based on Graph Neural Networks

This paper presents a new Intelligent Anti-Money Laundering Transaction Pattern Recognition System based on Graph Neural Networks (GNNs). The proposed system addresses the limitations of traditional anti-money laundering (AML) by leveraging the power of image representation and deep learning techniques. We introduce general methods for creating financial networks based on different shapes, including structural and physical. A custom GNN architecture is designed, featuring heterogeneous graph convolution, listening mechanisms, and physical models to capture the exchange patterns. The system uses advanced engineering techniques to extract both local and global features of financial performance. The analysis of the world's big data shows that the best performance of our method, achieved 35.2% Money Laundering Detection Rate (MLDR) in the top 1% of business flag, do better way. The model interpretation is improved by analyzing the SHAP value, providing insight into the decision-making process. Case studies show the system's ability to uncover financial transactions, including deposits from cryptocurrency exchanges and smurfing operations. This research contributes to the advancement of AML practices by introducing more accurate, flexible, and effective solutions for investigating financial crimes in complex financial systems more.

Developing Predictive Models for Financial Stability: Integrating Behavioral Analytics into Credit Risk Management

In this article, its author takes an attempt at investigating the use of behavioral analytics to complement credit risk management in the world of finance. Current mainstream credit risk models very much depend on conventional methods that incorporate historical financial information like credit scores and income while ignoring client behavioral contemporaneous patterns essential for modeling their risk levels. By this, the transactional patterns; payment behaviour and digital activities included besides the traditional financial parameters allows the financial institution to make superior models that gives better results on credit worthiness and risk identification earlier. It is crucial for both better risk assessment and for embracing more inclusion within lending networks and financial creativity. Analyzing the informational component of behavioral data as well as its applicability in credit risk modeling and considering the difficulties of implementing it in practice, the article provides an outline of the approaches to telemetry integration.

Geospatial data in AML risk management: A review of applications and best practices

This review explores the role of geospatial data in enhancing Anti-Money Laundering (AML) risk management, focusing on its applications, best practices, and strategic implications for organizations. Geospatial data, combined with advanced analytical tools, provides critical insights into transaction patterns, customer locations, and geographic risk factors, enabling financial institutions to detect suspicious activities more effectively. The paper examines how geospatial data integration supports compliance with regulatory frameworks, bolsters transaction monitoring systems, and facilitates enhanced due diligence processes. Through a systematic review of existing literature and case studies, this paper identifies key applications such as identifying high-risk jurisdictions, mapping transaction flows, and detecting cross-border money laundering schemes. Best practices are highlighted, including leveraging geospatial analytics in conjunction with machine learning algorithms, optimizing data governance, and ensuring interoperability across systems. The findings suggest that organizations adopting these practices are better equipped to combat money laundering by improving risk assessment, reducing false positives, and enhancing investigative efficiency. Additionally, the review discusses the challenges of data privacy, regulatory compliance, and the technical complexities associated with geospatial data integration. The paper concludes by outlining future prospects, recommending further advancements in real-time geospatial analytics, cross-sector collaborations, and the adoption of emerging technologies to strengthen AML frameworks. This review serves as a comprehensive guide for financial institutions seeking to incorporate geospatial data into their AML strategies, driving both compliance and operational efficiency.

Enhancing Banking Services through Data-Driven ATM Placement Strategies: A Case Study of PT Bank Rakyat Indonesia

This research optimizes the placement strategy for new Automated Teller Machines (ATMs) at PT Bank Rakyat Indonesia (BRI). The study employs a data-driven methodology that integrates data preparation, clustering techniques, and coefficient calculations to identify the most suitable ATM locations. This approach goes beyond static models by incorporating customer behavior and transaction data. The methodology utilizes a two-pronged approach. The first phase assesses the potential of various districts for ATM placement, considering factors such as demographics, transaction patterns, and competitor presence. K-means clustering is then employed to prioritize districts with the highest potential for benefiting from additional ATMs. The second phase involves calculating coefficients by analyzing correlations between existing ATM distribution and district potential scores. This analysis provides data-driven recommendations for the optimal number of new ATMs in each district. By leveraging robust methodological framework, this research offers valuable insights for strategic decision-making and resource allocation. The proposed approach can enhance banking service accessibility across diverse regions, improve customer satisfaction, and contribute to the optimization of BRI's ATM network.

Legal Support of Artificial Intelligence in Countering Anti-Money Laundering and Terrorism Financing Regimes in the BRICS Plus Countries

The increasing integration of artificial intelligence technologies into the financial structures of the BRICS Plus countries (comprising the original member countries of Brazil, Russia, India, China, and South Africa as well as the four new member countries of Egypt, Ethiopia, Iran, and the United Arab Emirates) presents both opportunities and challenges in combating economic crimes, which include money laundering and terrorism financing. This article explores the complex regulatory landscape that governs the application of artificial intelligence in these efforts. It examines how artificial intelligence can enhance the performance of anti-money laundering and counter-terrorism financing frameworks by enabling the evaluation of massive datasets, the identification of anomalous transaction patterns, and the automation of compliance procedures. Simultaneously, the article addresses the highly challenging situations that arise when using artificial intelligence. For instance, these technologies can make it difficult to understand the fluctuation of illicit price ranges, thereby complicating efforts to determine their origins and destinations. Through a comparative analysis of the frameworks throughout the BRICS Plus countries, this research highlights the varying levels of regulatory readiness of these frameworks and proposes pathways for harmonizing artificial intelligence-driven economic security measures. The overarching goal of an artificial intelligence model is to enhance both the effectiveness and the integrity of the financial sectors in the BRICS Plus consortium, necessitating a collaborative approach to combating financial crimes in an increasing number of digital economies across the world.

ADVANCED FRAUD DETECTION IN CARD-BASED FINANCIAL SYSTEMS USING A BIDIRECTIONAL LSTM-GRU ENSEMBLE MODEL

This article addresses the challenges of fraud in card-based financial systems and proposes effective detection and prevention strategies. By leveraging recent data analytics and real-time monitoring, the study aims to enhance transaction security and integrity. The authors review existing fraud detection methodologies, emerging trends, and the evolving tactics of fraudsters, emphasizing the importance of collaboration among financial institutions, regulatory agencies, and technology providers. Our proposed solution is an ensemble model combining Bidirectional Gated Recurrent Unit (BiGRU) and Bidirectional Long Short-Term Memory (BiLSTM) networks, designed to capture complex transactional patterns more effectively. Comparative analysis of six machine learning classifiers—AdaBoost, Naïve Bayes, Decision Tree, Logistic Regression, Random Forest, and Voting—demonstrates that our BiLSTM-BiGRU ensemble model outperforms traditional methods, achieving a fraud detection performance score of 89.22%. This highlights the advanced deep learning model's superior ability to enhance the robustness and reliability of fraud detection systems.

DEEP LEARNING MODEL FOR DETECTING TERROR FINANCING PATTERNS IN FINANCIAL TRANSACTIONS

Terror financing remains a critical threat to global security, with illicit actors continually adapting their methods to evade detection. Traditional financial monitoring systems often struggle to identify the complex and covert patterns associated with terror-related transactions due to their reliance on predefined rules and statistical thresholds. This study introduces advanced deep learning models designed to detect terror financing patterns within vast datasets of financial transactions. We developed and evaluated several neural network architectures, including Convolutional Neural Networks (CNNs) and Long Short-Term Memory networks (LSTMs), to capture both spatial and temporal transaction features. The models were trained and tested on a dataset comprising anonymized financial transactions labeled for suspicious activities related to terror financing. Our deep learning models demonstrated superior performance over conventional machine learning approaches, achieving higher accuracy, precision, and recall in identifying suspicious transactions. Notably, the LSTM-based model excelled in detecting sequential transaction patterns indicative of layering and integration stages commonly used in terror financing. The results underscore the potential of deep learning techniques in enhancing the capabilities of financial institutions and regulatory bodies to combat terror financing. Implementing such models can lead to more proactive and effective monitoring systems that adapt to evolving illicit financing strategies.

Detection of Anomalous Bitcoin Transactions in Blockchain Using ML

An Internet of Things (IoT)-enabled blockchain helps to ensure quick and efficient immutable transactions. Low-power IoT integration with the Bitcoin network has created new opportunities and difficulties for blockchain transactions. Utilising data gathered from IoT-enabled devices, this study investigates the application of ML regression models to analyse and forecast Bitcoin transaction patterns. Several ML regression algorithms, including Lasso Regression, Gradient Boosting, Extreme Boosting, Extra Tree, and Random Forest Regression, are employed to build predictive models. These models are trained using historical Bitcoin transaction data to capture intricate relationships between various transaction parameters. To ensure model robustness and generalisation, cross-validation techniques and hyperparameter tuning are also applied. The empirical results show that the Bitcoin cost prediction of blockchain transactions in terms of time series. Additionally, it highlights the possibility of fusing block- chain analytics with IoT data streams, illuminating how new technologies might work together to enhance financial institutions.

Multi-class Bitcoin mixing service identification based on graph classification

Due to its anonymity and decentralization, Bitcoin has long been a haven for various illegal activities. Cyber-criminals generally legalize illicit funds by Bitcoin mixing services. Therefore, it is critical to investigate the mixing services in cryptocurrency anti-money laundering. Existing studies treat different mixing services as a class of suspicious Bitcoin entities. Furthermore, they are limited by relying on expert experience or needing to deal with large-scale networks. So far, multi-class mixing service identification has not been explored yet. It is challenging since mixing services share a similar procedure, presenting no sharp distinctions. However, mixing service identification facilitates the healthy development of Bitcoin, supports financial forensics for cryptocurrency regulation and legislation, and provides technical means for fine-grained blockchain supervision. This paper aims to achieve multi-class Bitcoin Mixing Service Identification with a Graph Classification (BMSI-GC) model. First, BMSI-GC constructs 2-hop ego networks (2-egonets) of mixing services based on their historical transactions. Second, it applies graph2vec, a graph classification model mainly used to calculate the similarity between graphs, to automatically extract address features from the constructed 2-egonets. Finally, it trains a multilayer perceptron classifier to perform classification based on the extracted features. BMSI-GC is flexible without handling the full-size network and handcrafting address features. Moreover, the differences in transaction patterns of mixing services reflected in the 2-egonets provide adequate information for identification. Our experimental study demonstrates that BMSI-GC performs excellently in multi-class Bitcoin mixing service identification, achieving an average identification F1-score of 95.08%.

MIDDLEMEN IN SEARCH EQUILIBRIUM WITH INTENSIVE AND EXTENSIVE MARGINS

AbstractWe study endogenous intermediation activity, the implied transaction pattern—direct trade, indirect trade, or both—and implications for efficiency. Related papers have agents exchanging indivisible goods or assets, capturing only extensive margins (trade frequency). To capture intensive margins, we incorporate divisibility representing quantity or quality. We characterize equilibrium, show how it depends on bargaining powers and costs, and how intensive margins matter—for example, middlemen with high bargaining power may charge more but offer higher quantity/quality, thus improving welfare. A tax‐subsidy policy is designed to achieve efficiency. A monetary version lets us compare money and middlemen and further discuss policy.

BiLSTM4DPS: An attention-based BiLSTM approach for detecting phishing scams in ethereum

With the burgeoning adoption of blockchain technology, cryptocurrencies have surged in popularity, becoming a focal point of global interest. Concurrently, the emergence of cryptocurrency phishing scams poses a significant threat to the financial security of the blockchain ecosystem, inflicting substantial economic damage on platforms and users alike. This study introduces an innovative approach leveraging an attention-augmented Bidirectional Long Short Term Memory Network (BiLSTM), termed BiLSTM4DPS, for the detection of phishing scams within the Ethereum network. We initiate by converting account transaction records into sequences, thereby extracting temporal and latent patterns of transactions. Subsequently, we integrate BiLSTM with multi-head attention mechanisms and masking techniques to construct a robust classification model aimed at identifying fraudulent accounts. Extensive experiments were conducted to assess the efficacy of BiLSTM4DPS, particularly under scenarios with limited account activity data. The results demonstrate that BiLSTM4DPS achieves remarkable predictive accuracy, surpassing existing state-of-the-art methods.

Cloud-Based Transaction Fraud Detection: An In-depth Analysis of ML Algorithms

Context: Cloud-based services are increasingly central in financial technology, enabling scalable and efficient transactions. However, they also heighten vulnerability to fraud, challenging the security of online financial activities. Traditional fraud detection struggles against sophisticated tactics, highlighting the need for advanced, cloud-compatible solutions. Objectives: This study assesses machine learning (ML) algorithms' ability to detect fraud in cloud environments, focusing on Logistic Regression (LR), Decision Trees (DT), Random Forest (RF), Support Vector Machines (SVM), and XGBoost (XGB). It uses a comprehensive dataset to determine which ML model best identifies fraudulent transactions, aiming to optimize these models for accuracy, precision, and efficiency in real-time detection. Results: The XGBoost model outperformed others in fraud detection accuracy, with Random Forest and Decision Trees also showing high effectiveness. These models were particularly good at balancing precision and recall, minimizing false positives, and accurately identifying fraud in complex transaction patterns. Conclusion: ML, especially ensemble and boosting models like XGBoost and Random Forest, offers a strong approach to detecting transaction fraud in cloud-based financial systems. Their capacity to handle vast data volumes and adapt to new fraud patterns enhances cloud transaction security. Implication: The study provides a guide for financial and cloud services to implement sophisticated fraud detection systems. It emphasizes the importance of continual ML innovation to tackle digital finance fraud, suggesting that adopting these advanced ML models can significantly reduce transaction fraud risks, ensuring a secure, efficient, and trustworthy platform for users.

A Stealthy Communication Model with Blockchain Smart Contract for Bidding Systems

With the widespread adoption of blockchain technology, its public ledger characteristic enhances transaction transparency but also amplifies the risk of privacy breaches. Attackers can infer users’ real identities and behaviors by analyzing public transaction patterns and address relationships, posing a severe threat to users’ privacy and security, and thus hindering further advancements in blockchain applications. To address this challenge, covert communication has emerged as an effective strategy for safeguarding the privacy of blockchain users and preventing information leakage. But existing blockchain-based covert communication schemes rely solely on the immutability of blockchain itself for robustness and suffer from low transmission efficiency. To tackle these issues, this paper proposes a stealthy communication model with blockchain smart contract for bidding systems. The model initiates by preprocessing sensitive information using a secret-sharing algorithm-the Shamir (t, n) threshold scheme-and subsequently embeds this information into bidding amounts, facilitating the covert transfer of sensitive data. We implemented and deployed this model on the Ethereum platform and conducted comprehensive performance evaluations. To assess the stealthiness of our approach, we employed a suite of statistical tests including the CDF, the Kolmogorov–Smirnov test, Welch’s t-test and K–L divergence. These analyses confirmed that amounts carrying concealed information were statistically indistinguishable from regular transactions, thus validating the effectiveness of our solution in maintaining the anonymity and confidentiality of information transmission within the blockchain ecosystem.

Unlocking Blockchain UTXO Transactional Patterns and Their Effect on Storage and Throughput Trade-Offs

Unlocking Blockchain UTXO Transactional Patterns and Their Effect on Storage and Throughput Trade-Offs

Analysis of Fraud Patterns in Islamic Banking Transactions: Strategies and Implementation of Prevention

The research aims to analyze patterns of fraud in Sharia banking transactions and identify effective prevention strategies. The method used is a qualitative approach with case study data collection. The research findings reveal various common patterns of fraud, such as identity theft, document forgery, transaction manipulation, and misappropriation of funds. Factors such as weaknesses in the transaction system, weak security policies, and lack of employee training influence the occurrence of fraud. Prevention strategies that can be implemented include early detection systems, strict security policies, employee training, and cooperation with authorities. It is hoped that the research findings can assist Sharia banks in enhancing transaction security, protecting customers, and minimizing fraud risks.

Money laundering prevention in the digital age: Leveraging graph databases for effective solutions

As financial transactions increasingly migrate to the digital realm, the challenge of preventing money laundering has become complex (Ramada, 2022). The research presented in this paper explores innovative approaches to combating money laundering in the digital age, focusing on the application of graph databases as a powerful tool for effective solutions. The study deep dives into the capabilities of graph databases in modeling intricate relationships within financial networks, conducting network analysis, and detecting anomalies in transaction patterns. By leveraging these capabilities, financial institutions can enhance customer due diligence, monitor transactions in real-time, and visualize complex networks to uncover hidden connections indicative of money laundering activities. The paper contains examination of case studies, regulatory compliance considerations, and the integration of graph databases into existing anti-money laundering frameworks. Ultimately, objective of this research has been to contribute to the evolving landscape of money laundering prevention strategies by highlighting the potential of graph databases as a key technology in the digital age.

K‐Fuse: Credit card fraud detection based on a classification method with a priori class partitioning and a novel feature selection strategy

AbstractOnline transactions have become the dominant and most popular form of online payment in today's digital economy. Due to the growing popularity of e‐commerce and the convenience it offers, both consumers and businesses are rapidly adopting online transactions. Notably, credit cards have become one of the most popular and standard online payment methods. However, it should be noted that credit card transactions are not without challenges. In particular, detecting and preventing fraudulent transactions is a major concern of the online payment system. It is difficult to find an effective detection model that can detect the new patterns created by fraudsters, due to the constant evolution of their methods to exploit the vulnerability of current security protocols. These fraud patterns are evolving and may not correspond to existing documented models, leading to a reduction in their identification. In addition, the customer's behavior can affect the model detection as it is susceptible to change based on factors such as economic conditions, trends, and individual circumstances. When consumers deviate from their typical behavior, the model may generate false alerts, thereby reducing its ability to differentiate between legitimate and fraudulent transactions. This article presents a new supervised detection model, called K‐Fuse, which introduces an unsupervised phase in order to detect fraud patterns that may correspond to innovative models introduced by fraudsters. K‐Fuse is a supervised classification method that fuses three steps consisting of (i) unsupervised clustering to identify hidden patterns of transactions in a dataset, (ii) a novel feature selection criterion based on the unsupervised results, and (iii) supervised classification to exploit the results of clustering and feature selection to predict new transactions as fraudulent or legitimate.

Utilization of the FP-Growth Algorithm on MSME Transaction Data:Recommendations for Small Gifts from The Padang Region

The existence of adequate transaction data turns out to have a similar sales transaction pattern for MSMEs, so it would be a shame if it were left like that. Moreover, this data can be used to increase efficiency in MSMEs in the culinary sector, one of which is as a recommendation for small gifts. The study uses the Association Rules technique, whereas fp-growth is used to obtain a combination of elements. The goal is to facilitate MSMEs' ability to suggest small gifts to clients. The fp-growth algorithm calculation was implemented to process 2043 data originating from transaction data in MSMEs, with the specified minimum support value being 15%, while the minimum confidence value determined was 55%. The results of the trial obtained the two best rules, namely, "If a customer buys a list of small gifts from Balado Sanjai Chips, then the customer will buy Jangek Crackers" and "If a customer buys Jangek Crackers, then the customer will buy Sanjai Balado Chips".

Novel intelligent supervised neuro-structures for nonlinear financial crime differential systems

Artificial intelligence (AI)-based applications contribute to monitoring financial transactions and detect fraudulent activity in real-time by analyzing transaction patterns, consumer behavior, and other statistics, making them essential for quickly addressing potential threats in the fight against financial crime dynamics. Leveraging financial crime systems with intelligent supervised neuro-structures exploiting nonlinear autoregressive exogenous networks integrating damped least square (NARX-DLS) optimization methods to achieve an appropriate degree of accuracy and adaptability for the estimation of complex nonlinear financial crime differential systems (NFCDSs). The representative NFCDS for financial crime indicators is expressed as susceptible individuals, financial criminals, individuals under prosecution, imprisoned individuals, and honest individuals. The Adams numerical solver accomplishes the acquisition of synthetic data for the layer structure NARX-DLS algorithm execution to solve NFCDSs for various financial crime parameters, such as recruitment rate, influence rate, conversion rate to honest people, financial criminal prosecution rate per capita, discharge and acquittal rate from prosecutions, percentage of discharge rate from prosecution, transition rate to prison, and freedom rate. A sturdy overlap between the solutions of NARX-DLSs and the reference numerical results of NFCDSs implies that the error value is close to a desirable value of zero. The effectiveness of the NARX-DLSs is evidenced by including a variety of assessment metrics that carefully examine the model’s correctness and efficacy, including mean square error-based convergence arches, adaptive regulating parameters, error distribution, and input-error/cross-correlation analyses.