Transaction Dataset Research Articles

Real-Time Online Banking Fraud Detection Model by Unsupervised Learning Fusion

Digital trades and payments are becoming increasingly popular, as they typically entail monetary transactions. This not only makes electronic transactions more convenient for the end customer, but it also raises the likelihood of fraud. An adequate fraud detection system with a cutting-edge model is critical to minimizing fraud costs. Identifying fraud at the ideal time entails establishing and setting up ubiquitous systems to consume and analyze massive amounts of streaming data. Recent advances in data analytics methods and introducing open-source technology for big data storage and processing opened new options for detecting fraud. This study aims to tackle this critical issue by providing a newly real-time e-transaction fraud detection schema that consolidates the advantages of both unsupervised learners, including autoencoder and extended isolation forests, with cutting-edge big data gadgets such as Spark streaming and sparkling water. It addresses the shortage of non-fraudulent instances and handles the excessive dimension of the set of features. On two real-world transactional datasets, we assess our suggested technique. Compared with other current fraud identification systems, our methodology delivers an elevated accuracy yield of 99%. Furthermore, it outperforms state-of-the-art approaches in reliably identifying fraudulent samples. Doi: 10.28991/HIJ-2024-05-01-014 Full Text: PDF

Feature Selection Techniques for Enhancing Credit Card Fraud Detection Performance: A Hybrid Metaheuristic Approach Using Nature-Inspired Algorithms

Abstract: The surge in credit card usage has led to a parallel increase in fraudulent transactions, necessitating advanced detection systems. Traditional methods encounter challenges with high-dimensional and imbalanced datasets. This paper proposes a comprehensive approach integrating metaheuristic algorithms and deep learning techniques to enhance fraud detection accuracy. The first contribution, the Rock Hyrax Swarm Optimization Feature Selection (RHSOFS) algorithm, draws inspiration from the collective behaviors of rock hyrax swarms to effectively select relevant features from high-dimensional datasets. RHSOFS identifies an optimal subset of features critical for fraud detection through supervised machine learning. Complementing this, the second contribution leverages a hybrid deep learning model. Beginning by organizing transactional data and constructing a Logical Graph of Behavior Profile (LGBP) to abstract transaction details, the Modified Butterfly Optimization Algorithm (MBOA) selects important features from the dataset. The hybrid model, integrating Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN), establishes rational connections between transactional characteristics, enhancing detection performance. To evaluate the approach, comparative efficiency analyses against existing methods, including Differential Evolutionary Feature Selection (DEFS), Genetic Algorithm Feature Selection (GAFS), Particle Swarm Optimization Feature Selection (PSOFS), and Ant Colony Optimization Feature Selection (ACOFS), are conducted. Results demonstrate the superiority of the approach in terms of both reliability and recognition rate, validated through rigorous statistical testing. The proposed hybrid approach signifies a significant advancement in credit card fraud detection systems, offering enhanced accuracy and efficiency in combating fraudulent activities amidst the growing complexity of transactional datasets.

Impact of tropical storms on the banking sector in the British Colonial Caribbean

This paper investigates the impact of four historical tropical storms on the Colonial Bank’s operations in the British Caribbean between 1922 and 1927. By employing a high-frequency data set of bank transactions, this study reveals how these severe shocks influenced the banking activities of clients. The findings reveal a multifaceted and significant impact of tropical storm strikes on the banks’ operations, particularly a surge in borrowing via overdrafts of current accounts. Moreover, the study reveals the multifaceted nature of such storms’ impact on the bank’s functionality, with affected branches demonstrating an uptick in deposits and savings as a strategy to mitigate funding shocks. The results of the econometric analysis indicate that the impact of such storms on banks’ functionality during the early 20th century was significant and multidimensional. It highlights the critical role that the Colonial Bank plays in facilitating recovery from these devastating events and contributes to the existing literature by studying multiple shocks at different geographical locations and time frames.

An Empirical Investigation of Manufacturers’ Operations Innovations in New Product Development Enabled by E-Commerce Platforms

E-commerce platforms are playing an increasingly important role in influencing manufacturers’ supply chain and product decisions. An emerging supply chain innovation, known as the platform-based consumer-to-manufacturer (PC2M) model, has been initiated by several large e-commerce platforms based on established digital links between consumers and manufacturers. These links enable consumer inputs into manufacturers’ operations, indirectly by capturing consumer preferences from platform-accumulated big data and directly by enabling consumer interaction with manufacturers through the e-commerce platform. Although manufacturers are increasingly integrating PC2M into new product development (NPD), there is little research on operations innovations in connection with the PC2M model and its impact on manufacturers’ new product success. To fill this research gap, we investigate the PC2M model of JD.com, a leading e-commerce platform in China that launched the PC2M model in 2018. We first identify two uses of PC2M by manufacturers to facilitate product development—platform-enabled big data analytics (PBA) and platform-enabled simulated product trials (PST)—and explore how PC2M enables operations innovations in NPD. Next, drawing on the knowledge-based view, we develop research hypotheses and empirically examine whether PC2M adoption improves new product performance using a large-scale, transactional dataset from JD.com. Through a series of carefully executed analyses, our study consistently finds that use of either PBA or PST in manufacturers’ NPD processes improves new product performance. We also explore how these effects vary across product types and markets with varying new product introduction rates. The findings offer important managerial insights for improving new product success in today's data-rich environment.

Extending application of explainable artificial intelligence for managers in financial organizations

AbstractAnomalies are a significant challenge for businesses in all industries. Artificial intelligence (AI) based machine learning (ML) detection models can help find aberrant customer transaction behaviour in financial datasets. However, the output responses provided by these AI-based ML models lack transparency and interpretability, making it challenging for financial managers to comprehend the reasoning underlying the AI detections. Suppose managers cannot comprehend how and why AI models develop responses based on the input information. In such cases, AI is unlikely to enhance data-driven decision-making and add value to organizations. This article’s primary objective is to illustrate the capacity of the SHapley Additive exPlanations (SHAP) technique to give finance managers an intuitive explanation of the anomaly detections AI-based ML models generate for a specific customer transaction dataset. Theoretically, we contribute to the literature on international finance by offering a conceptual review of AI algorithmic explainability. We discuss its implications for sustaining a competitive advantage using the concepts of action design research methodology following the research onion framework. We also suggest an explainable AI implementation methodology based on SHAP as a valuable guide for finance managers seeking to boost the transparency of AI-based ML models and to alleviate trust difficulties in data-driven decision-making.

Institutional investors in the Portuguese credit market (1550–1800): The case of the Misericórdias

AbstractThis article questions the drivers behind the distribution of savings in different capital markets in Portugal between 1550 and 1800. A novel dataset of credit transactions, interest rates and debt service documents a shift in the lenders' investment behaviour. By 1712, one of the leading institutional creditors—the Misericórdias—had ceased to allocate funds to the sovereign debt market. Data reveal that this disinvestment was neither related to the poor performance of debt service nor to the lure of potentially higher returns on private credit. We argue that changes in the rationales for issuing debt justify the drop in the number of institutional investors in the public credit market, and this correlates with the heavy allocation of funds into private lending.

Federated learning model for credit card fraud detection with data balancing techniques

In recent years, credit card transaction fraud has resulted in massive losses for both consumers and banks. Subsequently, both cardholders and banks need a strong fraud detection system to reduce cardholder losses. Credit card fraud detection (CCFD) is an important method of fraud prevention. However, there are many challenges in developing an ideal fraud detection system for banks. First off, due to data security and privacy concerns, various banks and other financial institutions are typically not permitted to exchange their transaction datasets. These issues make traditional systems find it difficult to learn and detect fraud depictions. Therefore, this paper proposes federated learning for CCFD over different frameworks (TensorFlow federated, PyTorch). Second, there is a significant imbalance in credit card transactions across all banks, with a small percentage of fraudulent transactions outweighing the majority of valid ones. In order to demonstrate the urgent need for a comprehensive investigation of class imbalance management techniques to develop a powerful model to identify fraudulent transactions, the dataset must be balanced. In order to address the issue of class imbalance, this study also seeks to give a comparative analysis of several individual and hybrid resampling techniques. In several experimental studies, the effectiveness of various resampling techniques in combination with classification approaches has been compared. In this study, it is found that the hybrid resampling methods perform well for machine learning classification models compared to deep learning classification models. The experimental results show that the best accuracy for the Random Forest (RF); Logistic Regression; K-Nearest Neighbors (KNN); Decision Tree (DT), and Gaussian Naive Bayes (NB) classifiers are 99,99%; 94,61%; 99.96%; 99,98%, and 91,47%, respectively. The comparative results show that the RF outperforms with high performance parameters (accuracy, recall, precision and f score) better than NB; RF; DT and KNN. RF achieve the minimum loss values with all resampling techniques, and the results, when utilizing the proposed models on the entire skewed dataset, achieved preferable outcomes to the unbalanced dataset. Furthermore, the PyTorch framework achieves higher prediction accuracy for the federated learning model than the TensorFlow federated framework but with more computational time.

Individual investor trading and stock returns after the Covid-19 pandemic: Evidence from Korea

This paper reveals a significant attenuation in the previously observed positive association between individual investors’ purchases and stock returns following the Covid-19 pandemic. To investigate these changes, I utilize a comprehensive dataset of daily stock transactions, categorized by investor groups, from the Korean main board market. By employing a difference-in-difference regression model, I find that stocks influenced by the influx of amateur investors or attention-based trading post-pandemic are the key drivers behind this transformation. These findings suggest that policymakers might need to consider implementing systematic liquidity provision programs, such as market-making activities, to fill the liquidity gap created by individual investors.

Herding in the cryptocurrency market: A transaction-level analysis

We contribute to the literature on herding in the cryptocurrency market by using a unique data set of trader transactions. Using popular metrics, we find significant evidence of herding, which is primarily driven by individuals mimicking their own past trades, given the sporadic nature of information as well as the ambiguity and anonymity inherent in this market. Herding is higher during bearish periods as traders react more similarly to negative news. We find evidence of intentional herding due to informational cascades in less liquid cryptocurrencies, where significant price movements may be interpreted as valuable information. Traders with larger accounts tend to mimic their own past trades. Mature traders trade similarly due to their lower tolerance for risk and experimentation. We find herding differentials among traders that arise due to the environment governing the local financial system in which they are located. Moreover, persistence in herding is lower compared to what has been reported in other markets due to the higher degree of ambiguity of cryptocurrencies and the individuals trading them. Finally, market factors such as volatility, have a significant effect on herding. Our results shed light on how trader characteristics and market factors impact an individual’s propensity to herd.

Thermal comfort and retail sales: A big data analysis of extreme temperature's impact on brick-and-mortar stores

The effect of weather on retail sales has long been of great interest to both the business and academic fields. This study investigated the impact of extreme temperatures on brick-and-mortar retail stores in Seoul, Korea. Using a comprehensive credit card transaction dataset, high-resolution weather data, and a semiparametric model, we found a significant increase in sales during extreme temperature events: 4% during heatwaves exceeding 35 °C and 11% during cold spells below −15 °C. This finding is supported by the thermal comfort hypothesis in retail sales, which suggests that consumers are driven to temperature-controlled indoor environments and are inclined to purchase products that provide thermal comfort, such as hot or cold beverages. As extreme weather events become more frequent owing to climate change, accurate sales forecasting during such conditions becomes crucial for retailers. Insights from our research enable retailers to better predict sales under extreme temperature conditions and to strategize accordingly, such as by highlighting thermal comfort products or ensuring optimal indoor temperatures with efficient air conditioning or heating systems.

FePARM: The Frequency-Patterned Associative Rule Mining Framework on Consumer Purchasing-Pattern for Online Shops

1Malasowe, Bridget Ogheneovo,, 2Okpako, Abugor Ejaita, 3 , 4Ejeh, Patrick Ogholuwarami, 5Ojugo, Arnold Adimabua & 6 1Department of Computer Science, University of Delta, Agbor, Nigeria 2Department of Cyber Security, University of Delta, Agbor, Nigeria 3Department of Cyber Security, Faculty of Computing, Delta State University of Science and Technology, Ozoro. Nigeria. 4Department of Computer Science, Dennis Osadebay University, Anwai-Asaba, Nigeria 5 Department of Computer Science, Federal University of Petroleum Resources Effurun, Nigeria 6 Department of Computer Science, Federal University of Petroleum Resources Effurun, Nigeria E-mails: 1bridget.malasowe@unidel.edu.ng; 2ejaita.okpako@unidel.edu.ng; 3okpormd@dsust.edu.ng, 4 patrick.ejeh@dou.edu.ng; 5ojugo.arnold@fupre.edu.ng, 6ako.rita@fupre.edu.ng ABSTRACT Transaction data often is a true presentation of consumers’ buying behavior, stored as a set of relational records, which properly harnessed via mining – can help businesses improve their sales volume as a decision support system. Managing such a system can pose many issues to biz such as feature evolution, concept evolution, concept drift, and infinite data length – and often makes it impractical to effectively store such big-data. To curb this, previous studies have assumed data to be stationary in using associative rule mining. This has deprived such systems of the flexibility and adaptiveness required to handle the dynamics of concept drift that characterizes transaction datasets. Our study thus proposes a basket frequent pattern growth trained associative rule mining model to handle large data. The dataset was retrieved from the Delta-Mall Asaba and consists of 556,000 transaction consumer records. The model consists of six-layers, and yields the best result with a 0.1 value for both the confidence and support level(s) at a 94% accuracy, sensitivity of 87%, and a specificity of 32% with a 20-second convergence and processing time. Keywords: Big-Data Analysis, Feature Evolution, Inventory Model, Market Basket Analysis, Businesses, Transactional Data Streams, Concept Drift CISDI Journal Reference Format Malasowe, B.O., Okpako, A.E., Okpor, M.D., Ejeh, P.O., Ojugo, A.A. & Ako, R.E. (2024): FePARM: The Frequency-Patterned Associative Rule Mining Framework on Consumer Purchasing-Pattern for Online Shops. Computing, Information Systems, Development Informatics & Allied Research Journal. Vol 15 No 2, Pp 15-28. dx.doi.org/10.22624/AIMS/CISDI/V15N2P2-1. Available online at www.isteams.net/cisdijournal

FePARM: The Frequency-Patterned Associative Rule Mining Framework on Consumer Purchasing-Pattern for Online Shops

Transaction data often is a true presentation of consumers’ buying behavior, stored as a set of relational records, which properly harnessed via mining – can help businesses improve their sales volume as a decision support system. Managing such a system can pose many issues to biz such as feature evolution, concept evolution, concept drift, and infinite data length – and often makes it impractical to effectively store such big-data. To curb this, previous studies have assumed data to be stationary in using associative rule mining. This has deprived such systems of the flexibility and adaptiveness required to handle the dynamics of concept drift that characterizes transaction datasets. Our study thus proposes a basket frequent pattern growth trained associative rule mining model to handle large data. The dataset was retrieved from the Delta-Mall Asaba and consists of 556,000 transaction consumer records. The model consists of six-layers, and yields the best result with a 0.1 value for both the confidence and support level(s) at a 94% accuracy, sensitivity of 87%, and a specificity of 32% with a 20-second convergence and processing time. Keywords: Big-Data Analysis, Feature Evolution, Inventory Model, Market Basket Analysis, Businesses, Transactional Data Streams, Concept Drift CISDI Journal Reference Format Malasowe, B.O., Okpako, E.A., Ashioba, N.O., Ejeh, P.O., Ojugo, A.A. & Ako, R.E. (2024): FePARM: The Frequency-Patterned Associative Rule Mining Framework on Consumer Purchasing-Pattern for Online Shops. Computing, Information Systems, Development Informatics & Allied Research Journal. Vol 15 No 2, Pp 1-14. dx.doi.org/10.22624/AIMS/CISDI/V15N2P2. Available online at www.isteams.net/cisdijournal

Advancing financial fraud detection: Self-attention generative adversarial networks for precise and effective identification

The burgeoning prominence of electronic payment methods has significantly intensified the importance of financial fraud detection, especially concerning credit card fraud. The challenges encompass data imbalance, overfitting, and a limited capacity to realistically emulate the complexity and dynamics of fraudulent activities. This paper introduces self-attention generative adversarial networks (SAGANs) for the detection of credit card fraud. Leveraging self-attention mechanisms, SAGANs can distinguish salient features and patterns within extensive transaction datasets, thereby fostering a more profound understanding and refined identification of credit card fraud. By incorporating generative adversarial networks (GANs), our model is capable of generating data that mirrors actual fraudulent behavior, which consequently enhances and optimizes fraud detection algorithms. Empirical findings from numerous credit card transaction datasets underscore the considerable superiority of our approach in terms of accuracy and recall. Our proposed SAGANs more competently emulate the complexity and diversity inherent in fraudulent behavior, thereby augmenting both the precision and efficacy of fraud detection.

Scheduling Most Valuable Committees for the Sharded Blockchain

In a sharded blockchain, transactions are processed by a number of parallel committees. Thus, the transaction throughput can be largely boosted. A problem is that some groups of blockchain nodes consume large latency to form committees at the beginning of each epoch. Moreover, the heterogeneous processing capabilities of different committees also result in imbalanced consensus latency. Such imbalanced two-phase latency brings a large cumulative age to the transactions pending in transaction pool. Consequently, the blockchain throughput can be significantly degraded. We believe that a good committee-scheduling strategy can reduce the cumulative age of transactions, and thus benefit the throughput. However, we have not yet found a committee-scheduling mechanism that works for accelerating block formation in the context of blockchain sharding. To this end, this paper studies a fine-balanced tradeoff between the transactions’ throughput and their cumulative age in a large-scale sharded blockchain. We formulate this tradeoff as a utility-maximization problem, which is proved NP-hard. To solve this problem, we propose an online distributed Stochastic-Exploration (SE) algorithm, which guarantees a near-optimal system utility. We then rigorously analyze three theoretical properties of the proposed algorithm, including the theoretical convergence time, the probability of committees’ failure due to Sybil attacks, as well as the performance perturbation brought by committees’ offline events. Finally, we evaluate the proposed algorithm using the dataset of real-world blockchain transactions. The simulation results demonstrate that the proposed SE algorithm outperforms other baselines in terms of system utility, the valuable degree of yielded solutions, latency, and throughput performance.

Typological differentiation and time-series effects of urban renewal on housing prices

Urban renewal is a critical avenue and strategic tool for achieving high-quality and sustainable development, and the residential market plays a vital role in promoting urban renewal and urban growth. However, the impact of urban renewal on housing prices remains inconclusive, closely tied to the characteristics of the study area, the sources of research data, and so on. To address this, we employ the difference-in-difference method and a mixed panel dataset of housing transactions from 2010 to 2021 to analyze in detail the effects of different urban renewal types (Industrial Renewal, Residential Renewal, and Commercial Renewal) on surrounding housing prices in Bao'an District at different renewal phases (announcement, start, and completion). We found that Industrial Renewal exerts the most significant impact on housing prices at the beginning and completion of renewal, followed by Residential Renewal and Commercial Renewal. Moreover, Urban renewal generates both positive and negative externalities on housing prices, especially in the case of numerous small-scale renewal projects. The findings suggest that urban renewal should be supported by systematic research on renewal types, the establishment of scale entry threshold, and the monitoring of the entire process.

Towards Hyper-Relevance in Marketing: Development of a Hybrid Cold-Start Recommender System

Recommender systems position themselves as powerful tools in the support of relevance and personalization, presenting remarkable potential in the area of marketing. The cold-start customer problematic presents a challenge within this topic, leading to the need of distinguishing user features and preferences based on a restricted set of transactional information. This paper proposes a hybrid recommender system that aims to leverage transactional and portfolio information as indicating characteristics of customer behaviour. Four independent systems are combined through a parallelised weighted hybrid design. The first individual system utilises the price, target age, and brand of each product to develop a content-based recommender system, identifying item similarities. Secondly, a keyword-based content system uses product titles and descriptions to identify related groups of items. The third system utilises transactional data, defining similarity between products based on purchasing patterns, categorised as a collaborative model. The fourth system distinguishes itself from the previous approaches by leveraging association rules, using transactional information to establish antecedent and precedence relationships between items through a market basket analysis. Two datasets were analysed: product portfolio and transactional datasets. The product portfolio had 17,118 unique products and the included 4,408,825 instances from 2 June 2021 until 2 June 2022. Although the collaborative system demonstrated the best evaluation metrics when comparing all systems individually, the hybridisation of the four systems surpassed each of the individual systems in performance, with a 8.9% hit rate, 6.6% portfolio coverage, and with closer targeting of customer preferences and smaller bias.

A scalable and flexible basket analysis system for big transaction data in Spark

Basket analysis is a prevailing technique to help retailers uncover patterns and associations of sold products in customer shopping transactions. However, as the size of transaction databases grows, the traditional basket analysis techniques and systems become less effective because of two issues in the applications of the big data age: data scalability and flexibility to adapt different application tasks. This paper proposes a scalable distributed frequent itemset mining (ScaDistFIM) algorithm for basket analysis on big transaction data to solve these two problems. ScaDistFIM is performed in two stages. The first stage uses the FP-Growth algorithm to compute the local frequent itemsets from each random subset of the distributed transaction dataset, and all random subsets are computed in parallel. The second stage uses an approximation method to aggregate all local frequent itemsets to the final approximate set of frequent itemsets where the support values of the frequent itemsets are estimated. We further elaborate on implementing the ScaDistFIM algorithm and a flexible basket analysis system using Spark SQL queries to demonstrate the system’s flexibility in real applications. The experiment results on synthetic and real-world transaction datasets demonstrate that compared to the Spark FP-Growth algorithm, the ScaDistFIM algorithm can achieve time savings of at least 90% while ensuring nearly 100% accuracy. Hence, the ScaDistFIM algorithm exhibits superior scalability. On dataset GenD with 1 billion records, the ScaDistFIM algorithm requires only 360 s to achieve 100% precision and recall. In contrast, due to memory limitations, Spark FP-Growth cannot complete the computation task.

The metropolitan and neighborhood geographies of REIT- and private equity–owned single-family rentals

ABSTRACT After the U.S. foreclosure crisis, institutional investors purchased thousands of homes and converted them into rental properties. Past research links these entities to a number of negative outcomes, including rent increases and eviction filings. This paper examines a larger variety of investors, including private equity firms and contract sellers. Using a national dataset of real estate transactions from 2010 to mid-2017, we examine the inter- and intra-metropolitan geography of institutional investors. Consistent with prior research, we find that large publicly traded entities purchased homes in growing Sunbelt metros, yet some specific firms target weaker-market metros. Large, publicly traded firms have concentrated investment in higher-value neighborhoods with larger shares of white residents. In contrast, private equity firms and contract sellers tend to invest in relatively lower-value neighborhoods with larger shares of Black residents. Results suggest different potential implications for these diverse actors’ investments, from crowding out prospective homebuyers to racial targeting.

High-speed rail and distribution of economic activities: Evidence from prefecture-level cities in China

Creating new subcenters around suburban transport nodes is a popular policy choice for urban development. In this study, we investigate how high-speed rail stations reshape economics activities in less populous cities. Employing night-light data at the pixel level, we demonstrate that the establishment of a new station has resulted in a flattened economic density gradient for Chinese prefecture-level cities. Notably, however, the gravitational core of urban economic activity does not exhibit a discernible shift towards the station area. This evidence suggests that the station area has not formed into a new subcenter; instead, it might have contributed to urban sprawl. We show that the formation of a new subcenter is tied to several pivotal determinants, which include the level of urban agglomeration, the distance to the original urban center, and government investment in the station area. Further analysis, grounded in a comprehensive dataset of land transactions, reveals that there has been no significant land price appreciation within the station area. This finding raises questions regarding the effectiveness of government-led new town development around the station area.

The Use of Machine Learning to Detect Financial Transaction Fraud: Multiple Benford Law Model for Auditors

Background: Fraud in financial transaction is at the root of corruption issues recorded in organization. Detecting fraud practices has become increasingly complex and challenging. As a result, auditors require precise analytical tools for fraud detection. Grouping financial transaction data using K-Means Clustering algorithm can enhance the efficiency of applying Benford Law for optimal fraud detection. Objective: This study aimed to introduce Multiple Benford Law Model for the analysis of data to show potential concealed fraud in the audited organization financial transaction. The data was categorized into low, medium, and high transaction values using K-Means Clustering algorithm. Subsequently, it was reanalyzed through Multiple Benford Law Model in a specialized fraud analysis tool. Methods: In this study, the experimental procedures of Multiple Benford Law Model designed for public sector organizations were applied. The analysis of suspected fraud generated by the toolkit was compared with the actual conditions reported in audit report. The financial transaction dataset was prepared and grouped into three distinct clusters using the Euclidean distance equation. Data in these clusters was analyzed using Benford Law, comparing the frequency of the first digit’s occurrence to the expected frequency based on Benford Law. Significant deviations exceeding ±5% were considered potential areas for further scrutiny in audit. Furthermore, the analysis were validated by cross-referencing the result with the findings presented in the authorized audit organization report. Results: Multiple Benford Law Model developed was incorporated into an audit toolkit to automated calculations based on Benford Law. Furthermore, the datasets were categorized using K-Means Clustering algorithm into three clusters representing low, medium, and high-value transaction data. Results from the application of Benford Law showed a 40.00% potential for fraud detection. However, when using Multiple Benford Law Model and dividing the data into three clusters, fraud detection accuracy increased to 93.33%. The comparative results in audit report indicated a 75.00% consistency with the actual events or facts discovered. Conclusion: The use of Multiple Benford Law Model in audit toolkit substantially improved the accuracy of detecting potential fraud in financial transaction. Validation through audit report showed the conformity between the identified fraud practices and the detected financial transaction. Keywords: Fraud Detection, Benford’s Law, K-Means Clustering, Audit Toolkit, Fraudulent Practices.