Intelligent financial fraud detection

TL;DR: This work proposes a multitask learning framework, MultiFraud, for complex fraud detection with reasonable explanation, which enables multiple domains to share embeddings and enhance modeling capabilities for fraud detection.

TL;DR: This paper discusses two approaches of detecting the different anomalies types using basic sequential analysis techniques, including the classical one-dimensional approach and a simple approach to use multiple dimensions of the process information in the sequential analysis.

TL;DR: The paper examines the research on ML and data mining techniques as published in the past decade, identifying research gaps, recent techniques, and highlighting the increasing utilization of artificial neural networks in fraud detection within the industry.

TL;DR: The study aims are to survey the technical aspects of anti-money laundering systems (AML), review the existing machine learning algorithms and techniques applied to detect money laundering patterns, detect unusual behavior and money laundering groups, and pinpoint the study contribution in detecting moneyaundering groups.

TL;DR: This work describes the tools available for statistical fraud detection and the areas in which fraud detection technologies are most used, and statistics and machine learning provide effective technologies for fraud detection.

TL;DR: The main data mining techniques used for FFD are logistic models, neural networks, the Bayesian belief network, and decision trees, all of which provide primary solutions to the problems inherent in the detection and classification of fraudulent data.

TL;DR: In this paper, a profile of a sample of earnings manipulators, their distinguishing characteristics, and a suggested model for detecting manipulation is presented; the model's variables are designed to capture either the financial statement distortions that can result from manipulation or preconditions that might prompt companies to engage in such activity.

TL;DR: Among the six data mining techniques, artificial neural network is the only one that can accurately estimate the real probability of default, and its regression intercept is close to zero, and regression coefficient to one.