Search Results (68)

The increasing complexity of financial reporting has enabled the implementation of innovative accounting practices that often obscure a company’s actual performance. This project seeks to uncover manipulative behaviours by constructing an anomaly detection model that utilises unsupervised machine learning techniques. We examined a dataset of 149,566 Slovak firms from 2016 to 2023, which included 12 financial parameters. Utilising TwoSteps and K-means clustering in IBM SPSS, we discerned patterns of normative financial activity and computed an abnormality index for each firm. Entities with the most significant deviation from cluster centroids were identified as suspicious. The model attained a silhouette score of 1.0, signifying outstanding clustering quality. We discovered a total of 231 anomalous firms, predominantly concentrated in sectors C (32.47%), G (13.42%), and L (7.36%). Our research indicates that anomaly-based models can markedly enhance the precision of fraud detection, especially in scenarios with scarce labelled data. The model integrates intricate data processing and delivers an exhaustive study of the regional and sectoral distribution of anomalies, thereby increasing its relevance in practical applications. Full article

In such a dynamic and increasingly digitalized financial sector, many sophisticated fraudulent and cybercrime activities continue to challenge conventional detection systems. This research study explores a decentralized anomaly detection framework using deep autoencoders, designed to meet the dual imperatives of fraud detection effectiveness and user data privacy. Instead of relying on centralized aggregation or data sharing, the proposed model simulates distributed training across multiple financial nodes, with each institution processing data locally and independently. The framework is evaluated using two real-world datasets, the Credit Card Fraud dataset and the NeurIPS 2022 Bank Account Fraud dataset. The research methodology applied robust preprocessing, the implementation of a compact autoencoder architecture, and a threshold-based anomaly detection strategy. Evaluation metrics, such as confusion matrices, receiver operating characteristic (ROC) curves, precision–recall (PR) curves, and reconstruction error distributions, are used to assess the model’s performance. Also, a threshold sensitivity analysis has been applied to explore detection trade-offs at varying levels of strictness. Although the model’s recall remains modest due to class imbalance, it demonstrates strong precision at higher thresholds, which demonstrates its utility in minimizing false positives. Overall, this research study is a practical and privacy-conscious approach to fraud detection that aligns with the operational realities of financial institutions and regulatory compliance toward scalability, privacy preservation, and interpretable fraud detection solutions suitable for real-world financial environments. Full article

Predictive maintenance in industrial settings increasingly demands systems capable of integrating heterogeneous data streams while balancing computational efficiency and contextual reasoning. This paper introduces a novel framework leveraging Large Language Models (LLMs) to address these challenges in compressor monitoring, demonstrating their potential to enhance anomaly detection accuracy and operational cost-effectiveness. We evaluate Qwen 2.5-32B against traditional machine learning models (ANN, CNN, LSTM), achieving superior recall (92.3%) and AUC-ROC (0.991) through transformer-based architectures optimized for multimodal data fusion. A financial case study reveals operational cost reductions of 18% via reduced downtime and optimized maintenance schedules, while a real-time monitoring dashboard validates scalability for industrial deployment. Our findings highlight the transformative role of LLMs in bridging technical innovation with domain-specific operational constraints, offering a blueprint for predictive maintenance in niche industries. Full article

Stuck pipe events remain a critical challenge in oil and gas drilling operations, leading to increased non-productive time and substantial financial losses. Traditional detection methods rely on manual monitoring and expert judgment, which are prone to delays and human error. This study proposes a deep learning autoencoder-based anomaly diagnosis approach to enhance the detection of stuck pipe incidents. Using high-resolution time series drilling data from the Volve field, a deep learning autoencoder model was trained exclusively on normal drilling conditions to learn operational patterns and detect deviations indicative of stuck pipe events. The proposed model leverages reconstruction error as an anomaly detection metric, effectively distinguishing between normal and stuck cases. The results demonstrate that the model achieves a detection accuracy of 99.06%, with an area under the receiver operating characteristic curve (AUC) of 0.958. Additionally, the model attained a precision of 97.12%, a recall of 91.34%, and a F1-score of 94.15%, significantly reducing false positives and false negatives. The findings highlight the potential of deep learning-based approaches in improving real-time anomaly detection, offering a scalable and cost-effective solution for mitigating drilling disruptions. This research contributes to advancing intelligent monitoring systems in the oil and gas industry, reducing operational risks, and enhancing drilling efficiency. Full article

Robust credit risk prediction in emerging economies increasingly demands the integration of external factors (EFs) beyond borrowers’ control. This study introduces a scenario-based methodology to incorporate EF—namely COVID-19 severity (mortality and confirmed cases), climate anomalies (temperature deviations, weather-induced road blockages), and social unrest—into machine learning (ML) models for credit delinquency prediction. The approach is grounded in a CRISP-DM framework, combining stationarity testing (Dickey–Fuller), causality analysis (Granger), and post hoc explainability (SHAP, LIME), along with performance evaluation via AUC, ACC, KS, and F1 metrics. The empirical analysis uses nearly 8.2 million records compiled from multiple sources, including 367,000 credit operations granted to individuals and microbusiness owners by a regulated Peruvian financial institution (FMOD) between January 2020 and September 2023. These data also include time series of delinquency by economic activity, external factor indicators (e.g., mortality, climate disruptions, and protest events), and their dynamic interactions assessed through Granger causality to evaluate both the intensity and propagation of external shocks. The results confirm that EF inclusion significantly enhances model performance and robustness. Time-lagged mortality (COVID MOV) emerges as the most powerful single predictor of delinquency, while compound crises (climate and unrest) further intensify default risk—particularly in portfolios without public support. Among the evaluated models, CNN and XGB consistently demonstrate superior adaptability, defined as their ability to maintain strong predictive performance across diverse stress scenarios—including pandemic, climate, and unrest contexts—and to dynamically adjust to varying input distributions and portfolio conditions. Post hoc analyses reveal that EF effects dynamically interact with borrower income, indebtedness, and behavioral traits. This study provides a scalable, explainable framework for integrating systemic shocks into credit risk modeling. The findings contribute to more informed, adaptive, and transparent lending decisions in volatile economic contexts, relevant to financial institutions, regulators, and risk practitioners in emerging markets. Full article

This study presents an optimization for a distributed machine learning framework to achieve credit card fraud detection scalability. Due to the growth in fraudulent activities, this research implements the PySpark-based processing of large-scale transaction datasets, integrating advanced machine learning models: Logistic Regression, Decision Trees, Random Forests, XGBoost, and CatBoost. These have been evaluated in terms of scalability, accuracy, and handling imbalanced datasets. Key findings: Among the most promising models for complex and imbalanced data, XGBoost and CatBoost promise close-to-ideal accuracy rates in fraudulent transaction detection. PySpark will be instrumental in scaling these systems to enable them to perform distributed processing, real-time analysis, and adaptive learning. This study further discusses challenges like overfitting, data access, and real-time implementation with potential solutions such as ensemble methods, intelligent sampling, and graph-based approaches. Future directions are underlined by deploying these frameworks in live transaction environments, leveraging continuous learning mechanisms, and integrating advanced anomaly detection techniques to handle evolving fraud patterns. The present research demonstrates the importance of distributed machine learning frameworks for developing robust, scalable, and efficient fraud detection systems, considering their significant impact on financial security and the overall financial ecosystem. Full article

Since the 2008 global economic crisis, the detection of financial instabilities has garnered extensive research attention, particularly through the application of time-series analysis. In this study, a novel time-series analysis method, integrating the Kullback–Leibler Divergence (KLD) metric with a sliding window technique, is proposed to detect instabilities in time-series data, especially in financial markets. Global financial time series from 2004 to 2022 were analyzed. The raw time series were preprocessed into return rate series and transformed into complex networks using the directed horizontal visibility graph (DHVG) algorithm, effectively preserving temporal variabilities in network topologies. The KLD method was evaluated through both retrospective analysis and real-time monitoring. It successfully identified idiosyncratic incidents in the financial market, correlating them with specific economic events. Compared to traditional metrics (e.g., moments) and econometric methods, KLD demonstrated superior performance in capturing sequence information and detecting anomalies without requiring linear regression models. Although initially designed for financial data, the KLD method is versatile and can be applied to other types of time series as well. Full article

The increasing cost of energy and the prevalence of electricity theft pose significant financial and operational challenges for energy providers. Traditional fraud detection methods often fail to identify sophisticated unauthorized consumption, particularly in non-smart-grid environments. This study proposes an advanced machine learning-based methodology for detecting energy fraud, leveraging real-world data from energy distribution networks. This approach integrates multiple machine learning models—k-nearest neighbors (kNN), decision trees, random forest, and artificial neural networks (ANNs)—to improve detection accuracy and efficiency. Experimental results demonstrate an 89.5% fraud detection accuracy, significantly outperforming conventional methods. Furthermore, the implementation of this model led to an estimated financial loss reduction of EUR 45,200. By analyzing historical consumption patterns, anomaly detection techniques, and geospatial data, the proposed system enhances fraud detection capabilities across both smart and non-smart grids. Future research will focus on real-time detection, scalability, and the integration of external data sources to further refine predictive accuracy. Full article

Detecting fraudulent activities such as Ponzi schemes within smart contract transactions is a critical challenge in decentralized finance. Existing methods often fail to capture the heterogeneous, multi-faceted nature of blockchain data, and many graph-based models overlook the contextual patterns that are vital for effective anomaly detection. In this paper, we propose MVCG-SPS, a Multi-View Contrastive Graph Neural Network designed to address these limitations. Our approach incorporates three key innovations: (1) Meta-Path-Based View Construction, which constructs multiple views of the data using meta-paths to capture different semantic relationships; (2) Reinforcement-Learning-Driven Multi-View Aggregation, which adaptively combines features from multiple views by optimizing aggregation weights through reinforcement learning; and (3) Multi-Scale Contrastive Learning, which aligns embeddings both within and across views to enhance representation robustness and improve anomaly detection performance. By leveraging a multi-view strategy, MVCG-SPS effectively integrates diverse perspectives to detect complex fraudulent behaviors in blockchain ecosystems. Extensive experiments on real-world Ethereum datasets demonstrated that MVCG-SPS consistently outperformed state-of-the-art baselines across multiple metrics, including F1 Score, AUPRC, and Rec@K. Our work provides a new direction for multi-view graph-based anomaly detection and offers valuable insights for improving security in decentralized financial systems. Full article

“Normal” events are characterized as data patterns or behaviors that align with expected operational conditions, while “anomalies” are defined as deviations from these patterns, potentially signaling faults, errors, or unexpected system behaviors. The timely and accurate detection of anomalies plays a critical role in domains such as industrial manufacturing, financial transactions, and other related domains. In the context of Industry 4.0, the proliferation of sensors has resulted in a massive influx of time series data, making the anomaly detection of such multivariate time series data a popular research area. Long Short-Term Memory (LSTM) has been extensively recognized as an effective framework for modeling and processing time series data. Previous studies have combined Bi-directional Long Short-Term Memory (Bi-LSTM) architecture with Autoencoder (AE) for multivariate time series anomaly detection. However, due to the inherent limitations of LSTM, Bi-LSTM-AE still cannot overcome these drawbacks. Our study replaces the LSTM units within the Bi-LSTM-AE architecture of existing research with Working Memory Connections for LSTM units and demonstrates that this architecture performs better in the field of multivariate time series anomaly detection compared to using standard LSTM units. The model we proposed not only outperforms the baseline models but also demonstrates greater robustness across various scenarios. Full article

This study presents a Generative AI-Enhanced Cybersecurity Framework designed to strengthen enterprise data privacy management while improving threat detection accuracy and scalability. By leveraging Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), and traditional anomaly detection methods, the framework generates synthetic datasets that mimic real-world data, ensuring privacy and regulatory compliance. At its core, the anomaly detection engine integrates machine learning models, such as Random Forest and Support Vector Machines (SVMs), alongside deep learning techniques like Long Short-Term Memory (LSTM) networks, delivering robust performance across diverse domains. Experimental results demonstrate the framework’s adaptability and high performance in the financial sector (accuracy: 94%, recall: 95%), healthcare (accuracy: 96%, precision: 93%), and smart city infrastructures (accuracy: 91%, F1 score: 90%). The framework achieves a balanced trade-off between accuracy (0.96) and computational efficiency (processing time: 1.5 s per transaction), making it ideal for real-time enterprise deployments. Unlike analog systems that achieve > 0.99 accuracy at the cost of higher resource consumption and limited scalability, this framework emphasizes practical applications in diverse sectors. Additionally, it employs differential privacy, encryption, and data masking to ensure data security while addressing modern cybersecurity challenges. Future work aims to enhance real-time scalability further and explore reinforcement learning to advance proactive threat mitigation measures. This research provides a scalable, adaptive, and practical solution for enterprise-level cybersecurity and data privacy management. Full article

Blockchains are the backbone behind cryptocurrency networks, which have developed rapidly in the last two decades. However, this growth has brought several challenges due to the features of these networks, specifically anonymity and decentralization. One of these challenges is the fight against fraudulent activities performed in these networks, which, among other things, involve financial schemes, phishing attacks or money laundering. This article will address the problem of identifying fraud cases among a large set of transactions extracted from the Bitcoin network. More specifically, our study’s goal was to find reliable techniques to label Bitcoin transactions, taking into account their features. The approach followed involved two kinds of Machine Learning methods. On the one hand, anomaly detection algorithms were applied to determine whether fraudulent activities tend to show anomalous behaviour without resorting to manually obtained labels. On the other hand, Heterogeneous Graph Transformers were used to leverage the heterogeneous relational nature of the cryptocurrency information. As a result, the article will provide reasonable conclusions to acknowledge that unsupervised approaches can be useful for fraud detection on blockchain networks. Furthermore, the effectiveness of supervised graph methods was revalidated, emphasizing the importance of data heterogeneity. Full article

Deteriorating road infrastructure is a global concern, especially in low-income countries where financial and technological constraints hinder effective monitoring and maintenance. Traditional methods, like inertial profilers, are expensive and complex, making them unsuitable for large-scale use. This paper explores the integration of cost-effective, scalable smartphone technologies for road surface monitoring. Smartphone sensors, such as accelerometers and gyroscopes, combined with data preprocessing techniques like filtering and reorientation, improve the quality of collected data. Machine learning algorithms, particularly CNNs, are utilized to classify road anomalies, enhancing detection accuracy and system efficiency. The results demonstrate that smartphone-based systems, paired with advanced data processing and machine learning, significantly reduce the cost and complexity of traditional road surveys. Future work could focus on improving sensor calibration, data synchronization, and machine learning models to handle diverse real-world conditions. These advancements will increase the accuracy and scalability of smartphone-based monitoring systems, particularly for urban areas requiring real-time data for rapid maintenance. Full article

Investigating the stability of and fluctuations in the energy market has long been of interest to researchers and financial market participants. This study aimed to analyze the Chinese energy market, focusing on its volatility and response to financial tensions. For this purpose, data from eight major financial companies, which were selected based on their market share in Shanghai’s and Shenzhen’s financial markets, were collected from January 2014 to December 2023. In this study, stock prices and trading volumes were used as the key variables to build bootstrap-based minimum spanning trees (BMSTs) using ratio mutual information (RMI). Then, using the sliding window procedure, the major network characteristics were derived to create an anomaly-detection tool using the multivariate exponentially weighted moving average (MEWMA), along with the Brent crude oil price index as a benchmark and a global oil price indicator. This framework’s stability was evaluated through stress testing with five scenarios designed for this purpose. The results demonstrate that during periods of high oil price volatility, such as during the turbulence in the stock market in 2015 and the COVID-19 pandemic in 2020, the network topologies became more centralized, which shows that the market’s instability increased. This framework successfully identifies anomalies and proves to be a valuable tool for market players and policymakers in evaluating companies that are active in the energy sector and predicting possible instabilities, which could be useful in monitoring financial markets and improving decision-making processes in the energy sector. In addition, the integration of other macroeconomic factors into this field could strengthen the identification of anomalies and be considered a field for possible research. Full article

Multidimensional enterprise performance characteristics (enterprise operational data, financial transactions records) are stored in the company’s database (warehouse), and their volume and variety are huge. Financial transaction data are directly and indirectly related to value chain processes, various physical objects of activity, and their attributes. There are data mining (DM) and process mining (PM) methods for analyzing enterprise operational data and identifying deficiencies in business process management. There is a need to find new user experience (UX)-driven methods for user interface with the specification of DM and PM tools on the level of business process management concepts. The paper presents the UX design-based approach to designing the user interface (UI) of process mining and data mining systems and is based on a conceptual semantic model named financial data space (FDS). The peculiarity of FDS is that it can include the characteristics of financial data and other UX-related characteristics (events, environmental and internal changes, business location) that may have an impact on changes in the values of financial objects (FO). The presented multidimensional financial data model helps increase the possibility of uncovering management weaknesses by identifying anomalies in large amounts of financial data. The prototypes of components of the financial data analysis system are described and developed using the process mining tool. The presented method of a multidimensional representation of financial data and transformation into a PM project is a user-friendly solution that allows to increase the analytical capabilities of the auditor’s work with large amounts of data, providing a more flexible view of the financial indicators of the company’s activity. Full article