ILoRE: Dynamic Graph Representation with Instant Long-term Modeling and Re-occurrence Preservation

DyHDGE: Dynamic heterogeneous transaction graph embedding for safety-centric fraud detection in financial scenarios

Dynamic graph representation learning with disentangled information bottleneck.

Dynamic graph representation has gradually attached more attention in the research area since many real-world graphs evolve over time. Recent works typically represent the dynamic graph as a sequence of discrete static subgraphs and apply a recurrent neural network (RNN) to capture graph evolution over multiple discrete-time subgraphs. However, these approaches fail to capture the fine-grained temporal evolution leading to introducing temporal invalid information into the structure representation. In this paper, we propose a novel temporal evolution-enhanced dynamic graph embedding method (TE-DyGE) to address this limitation. To capture the fine-grained temporal evolution, TE-DyGE first applies a temporal random walk to filter out temporally valid information. Additionally, we introduce temporal-dependent weight to enhance the structure attention network. We evaluate the TE-DyGE in link prediction tasks. The experimental results demonstrate a generally higher performance of TE-DyGE compared with several state-of-the-art related baselines.

We introduced a novel methodology Multivariate Degree Distribution to Dynamic Graph (MDD2DG) with Inter-channel Relevance Attention (IRA) mechanism to analyze multi-channel Electrocardiogram (ECG) signals and explore signal connections across different channels. Our methodology comprises three main steps. First, multi-channel cardiac signals are transformed into multi-channel visual graphs to extract crucial degree distribution features. Then, degree distributions are mapped into dynamic graphs using a neural network with an IRA mechanism. After that, critical features are extracted within dynamic graphs utilizing a Graph Convolutional Neural Networks (GCNNs), and classification is subsequently performed using a multilayer perceptron. In this model, a method of multi-scale position embedding was introduced, which significantly enhanced the processing efficiency of the model by providing a simpler yet sufficiently effective feature representation. Compared to traditional complex network methods, our approach replaces fixed formula-calculated features with dynamic graph models, resulting in improved recognition accuracy. In the experiments, we achieved an impressive 99.94% classification accuracy for distinguishing ECG signals from the five distinct locations (AMI, ASMI, ALMI, IMI and ILMI) with myocardial infarction (MI) as well as those of the healthy controls (HC). This work contributes to the analysis of complex physiological signals in the field of multi-channel ECG sequence, and provides a robust approach with promising implications for improving clinical medicine and the early detection of cardiac diseases.

Multi-Camera Multiple Object Tracking (MC-MOT) is a significant computer vision problem due to its emerging applicability in several real-world applications. Despite a large number of existing works, solving the data association problem in any MC-MOT pipeline is arguably one of the most challenging tasks. Developing a robust MC-MOT system, however, is still highly challenging due to many practical issues such as inconsistent lighting conditions, varying object movement patterns, or the trajectory occlusions of the objects between the cameras. To address these problems, this work, therefore, proposes a new Dynamic Graph Model with Link Prediction (DyGLIP) approach <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> to solve the data association task. Compared to existing methods, our new model offers several advantages, including better feature representations and the ability to recover from lost tracks during camera transitions. Moreover, our model works gracefully regardless of the overlapping ratios between the cameras. Experimental results show that we out-perform existing MC-MOT algorithms by a large margin on several practical datasets. Notably, our model works favor-ably on online settings but can be extended to an incremental approach for large-scale datasets.

The rapid advancement of artificial intelligence (AI) technologies, particularly generative AI and large language models (LLMs), has ushered in a new era of possibilities for the financial sector. This paper explores the integration of these cutting-edge technologies into financial sector risk management, examining both the potential applications and the necessary regulatory frameworks. We provide a comprehensive analysis of how generative AI and LLMs can revolutionize risk assessment, fraud detection, market analysis, and regulatory compliance. The study delves into the technical aspects of these AI models, their implementation challenges, and the implications for existing risk management practices. Furthermore, we propose a novel framework for the responsible adoption of AI in financial risk management, addressing concerns related to model interpretability, data privacy, and algorithmic bias. Our findings suggest that while generative AI and LLMs offer unprecedented opportunities for enhancing risk management capabilities, they also necessitate a recalibration of regulatory approaches to ensure financial stability and consumer protection. This research contributes to the growing body of literature on AI in finance and provides actionable insights for practitioners, policymakers, and researchers in the field.

Dynamic networks require effective methods of monitoring and surveillance in order to respond promptly to unusual disturbances. In many applications, it is of interest to identify anomalous behavior within a dynamic interacting system. Such anomalous interactions are reflected by structural changes in the network representation of the system. In this paper, a dynamic random graph model is proposed that takes into account the past activities of the individuals in the social network and also represents temporal dependency of the network. The model parameters are appearance and disappearance probabilities of an edge which are estimated using a maximum likelihood approach. A generalization of a single path‐dependent likelihood ratio test is employed to detect changes in the parameters of the proposed model. Through monitoring the estimated parameters, one can effectively detect structural changes in a temporal‐dependent network. The proposed model is employed to describe the behavior of a real network, and its parameters are monitored via dependent likelihood ratio test and multivariate exponentially weighted moving average control chart. Results indicate that the proposed dynamic random graph model is a reliable mean to modeling and detecting changes in temporally dependent networks.

The purpose of this study is to respond to research gaps and inconsistencies in research findings by empirically testing how financial targets and external pressures affect the detection of financial statement fraud. The study utilizes a descriptive research design with a quantitative approach. A method of purposive sampling was used, focusing on companies listed on the Jakarta Islamic Index (JII) from 2019 to 2023. Statistical analysis was done using multiple regression analysis. The results show that financial targets have a significant effect on the detection of financial statement fraud, while external pressures do not have a significant effect. However, both financial targets and external pressures have a simultaneous effect on the detection of financial statement fraud. The practical implications, financial targets, and external pressures can have an impact on the detection of fraud within the organization. Under these conditions, fraud may be more difficult to detect because the organization's or individual's focus is more on achieving targets or meeting external expectations. Therefore, companies need to have a strong internal control system and independent audits to help identify and manage potential risks of financial statement fraud.

In the intelligent transportation system, traffic forecasting, which is generally characterized as a graph spatial-temporal prediction task, plays a crucial role. It is challenging to generate reliable forecast results due to the complexity of traffic topological information and the inherent uncertainty of road traffic circumstances. Existing works generally focus on modeling spatial dependency on static graph structures, but ignore dynamic relations between road segments and cannot extract long-range traffic dependencies in spatial-temporal domains. To bridge the above gaps, we present a novel framework, called <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Dynamic Multi-Hop Graph Attention Network</i> (DMGAN). Specifically, we leverage dynamic graph modeling to capture time-varying relations across road sections and introduce the multi-hop operation in each message propagation layer to extract long-range spatial dependency. Meanwhile, we develop a fusion-attention module, preserving both local and global hidden layer outputs of the encoder, to capture both long- and short-term temporal dependencies jointly. In this way, our method can fully model complex time-varying traffic topology information and capture the internal patterns of traffic series by integrating dynamic graph structure and temporal attention component. DGMAN achieves state-of-the-art performance in three metrics, as demonstrated by experimental findings on four real-world public traffic datasets, METR-LA, PEMS-BAY, PEMS03, and PEMS07. This code and data are available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/EEHITer/2022-TKDE-DMGAN-Pytorch/tree/main</uri> for reproducibility and further studies.

Financial fraud is one of the most crucial challenges in the digital economy, with growing cutting-edge attacks threatening the security of online transactions. Since conventional fraud detection models are often inadequate in identifying advanced fraud schemes, Advanced fraud detection approaches are required to prevent and recognize fraud in real time. In recent years, the progression of AI methods has received considerable attention in the financial sector, specifically in Financial Fraud Detection (FFD). Furthermore, Explainable AI (XAI) has become a prerequisite for building trust and driving acceptance of AI methods in high-stakes domains, namely credit risks, financial crime, and healthcare, which require reliability, fairness, and safety. This study presents an Optimization-Based Deep Belief Network for Fraud Detection in Financial Transaction Systems (ODBN-FDFTS) approach, aiming to develop a reliable system for detecting financial transaction fraud. The ODBN-FDFTS approach begins with data preprocessing using a standard scaler to normalize financial transaction records and enhance data quality. The Artificial Rabbit Optimization (ARO) technique is employed for Feature Selection (FS), and a Deep Belief Network (DBN) is used for the financial fraud classification process, tuning its hyperparameters using the Butterfly Optimization Algorithm (BOA). LIME is integrated to provide transparency and interpretability in the fraud detection process. The comparison study of the ODBN-FDFTS model showed a superior accuracy of 97.95% over other methods on the FFD dataset.

The purpose of this study was to empirically examine and accurately synthesise the effect of auditor quality, internal auditor professional ethics and internal auditor independence on the fraud detection process with information technology as a moderating variable. The process of detecting financial and economic fraud in a company plays an important role in the internal control system. The population of this study consists of internal auditors from companies located in Jakarta and Bogor. The results of this study were conducted using reliability and validity tests, with data analysis performed through descriptive analysis with three dependent variables: the Quality of Auditors as the first dependent variable (X1), the Professional Ethics as the second dependent variable (X2), and the Independence as the third dependent variable (X3). Fraud Detection serves as the independent variable (Y), while Information Technology acts as the moderating variable (Z). The results of this study indicate that the influence of auditor quality on fraud detection is positive and significant, the influence of professional ethics on fraud detection is positive and significant, the influence of independence on fraud detection is positive and significant, and the role of information technology in the influence of audit quality is positive but not significant, the role of information technology in the influence of professional ethics on fraud detection is positive but not significant, and the influence of independence on fraud detection through information technology as a moderating variable is positive and significan. This research can have implications for improving fraud prevention and detection by comprehensively improving audit quality, professional ethics and internal auditor independence by maximising the use of information technology Keywords: Auditor Quality, Profesional Ethics, Independenc and Information Technology

Optimising dynamic graphical models for video content analysis

With the fast growth of financial transaction fraud, there is a need for advanced detection systems capable of real-time analysis. Rule-based and machine-learning approaches to fraud traditionally suffer from being unable to adapt to changing fraud patterns, returning very high back result rates and much inefficiency in the security of financial operations. However, convolutional neural networks (CNNs) and recurrent neural networks (RNNs) methods are suitable, but they lack adaptability and interpretability. This paper proposes an Adaptive Transactional Anomaly Detection Network (ATAD-Net), a new deep learning (DL) framework for improving fraud detection accuracy, minimizing false positives, and guaranteeing real-time adaptability. ATAD-Net dynamically adjusts to evolving fraud tactics by integrating CNNs for local pattern recognition and Long Short-Term Memory (LSTM) for sequential transaction analysis. After training and testing the model using the IEEE CIS Credit Card Fraud Detection Dataset, a large-scale benchmark for evaluating financial fraud detection models, the accuracies of the different models were assessed.

This paper introduces a new class of Bayesian dynamic models for inference and forecasting in high-dimensional time series observed on networks. The new model, called the dynamic chain graph model, is suitable for multivariate time series which exhibit symmetries within subsets of series and a causal drive mechanism between these subsets. The model can accommodate high-dimensional, non-linear and non-normal time series and enables local and parallel computation by decomposing the multivariate problem into separate, simpler sub-problems of lower dimensions. The advantages of the new model are illustrated by forecasting traffic network flows and also modelling gene expression data from transcriptional networks.

Traffic prediction plays a crucial role in the efficient operation of satellite networks. However, due to resource consumption arising from redundant training of multiple individual prediction models, the dynamic and coupled spatial-temporal relationship of traffic, and maintenance of accurate traffic proportions, this problem is non-trivial to solve. Therefore, we consider this problem and makes the following contributions. First, a multi-granularity traffic prediction framework based on a shared feature extraction is designed to jointly predict total network traffic and service-specific traffic of satellite networks. This design ensures that both global and per service predictions benefit from common representations, reduces redundant computations and lowers overall model complexity. Second, a dynamic adaptive graph with Graph Diffusion Convolution (GDC) and Gated Recurrent Units (GRUs) is proposed to extract the spatial-temporal dependency of network traffic by fusing the features of population coverage, satellite distances and historical traffic data. Third, to preserve the proportional relationship of the network traffic, the angle-based loss is employed to minimize the angle deviation between the predicted and truth traffic vectors. Meanwhile, a multi-task loss function is proposed that jointly optimizes the total traffic prediction loss, the service-level losses, and the consistency regularization term to achieve accurate multi-granularity prediction. Numerical results demonstrate that the proposed framework can reduce prediction error and improve correlation for both global and service-level predictions.