Improving Fraud Detection: Evaluating Auditors' Reactions to Abnormal Inconsistencies between Financial and Non-Financial Measures?

Fraud has become a worldwide phenomenon and prime issue of concern. It dwells in all countries and affects all types of organizations irrespective of their size, profitability or industry. The primary objective of this paper is to provide an in-depth understanding of literature related to corporate fraud in order to understand why fraud occurs and how to combat it. Research studies published during the period commencing from the year 1984 to 2014 have been reviewed. The study aims to provide an in-depth discussion on significant red flags that may exist before fraud occurrence. It, also, provides a comprehensive view about fraud detection and prevention methods. Findings reveal that red flag is an important mechanism to prevent fraud. Application of single fraud detection technique will not curb the fraud effectively. Also, the top executives were found to be responsible for implementing anti-fraud policies and techniques within business organization. Further, the present study tries to discern the research gap in existing literature and explore the area of future research.

Download Full-text

Fraud Detection Protocol for Web-Based Research Among Men Who Have Sex With Men: Development and Descriptive Evaluation (Preprint)

10.2196/preprints.12344 ◽

2018 ◽

Author(s):

April M Ballard ◽

Trey Cardwell ◽

April M Young

Keyword(s):

Survey Research ◽

Hiv Transmission ◽

Personal Information ◽

Fraud Detection ◽

Detection Algorithm ◽

Cross Sectional ◽

Web Based ◽

Young Msm ◽

Data Survey ◽

Sex With Men

BACKGROUND Internet is becoming an increasingly common tool for survey research, particularly among “hidden” or vulnerable populations, such as men who have sex with men (MSM). Web-based research has many advantages for participants and researchers, but fraud can present a significant threat to data integrity. OBJECTIVE The purpose of this analysis was to evaluate fraud detection strategies in a Web-based survey of young MSM and describe new protocols to improve fraud detection in Web-based survey research. METHODS This study involved a cross-sectional Web-based survey that examined individual- and network-level risk factors for HIV transmission and substance use among young MSM residing in 15 counties in Central Kentucky. Each survey entry, which was at least 50% complete, was evaluated by the study staff for fraud using an algorithm involving 8 criteria based on a combination of geolocation data, survey data, and personal information. Entries were classified as fraudulent, potentially fraudulent, or valid. Descriptive analyses were performed to describe each fraud detection criterion among entries. RESULTS Of the 414 survey entries, the final categorization resulted in 119 (28.7%) entries identified as fraud, 42 (10.1%) as potential fraud, and 253 (61.1%) as valid. Geolocation outside of the study area (164/414, 39.6%) was the most frequently violated criterion. However, 33.3% (82/246) of the entries that had ineligible geolocations belonged to participants who were in eligible locations (as verified by their request to mail payment to an address within the study area or participation at a local event). The second most frequently violated criterion was an invalid phone number (94/414, 22.7%), followed by mismatching names within an entry (43/414, 10.4%) and unusual email addresses (37/414, 8.9%). Less than 5% (18/414) of the entries had some combination of personal information items matching that of a previous entry. CONCLUSIONS This study suggests that researchers conducting Web-based surveys of MSM should be vigilant about the potential for fraud. Researchers should have a fraud detection algorithm in place prior to data collection and should not rely on the Internet Protocol (IP) address or geolocation alone, but should rather use a combination of indicators.

Download Full-text

Tax Fraud Detection through Neural Networks: An Application Using a Sample of Personal Income Taxpayers

Future Internet ◽

10.3390/fi11040086 ◽

2019 ◽

Vol 11 (4) ◽

pp. 86 ◽

Cited By ~ 2

Author(s):

César Pérez López ◽

María Delgado Rodríguez ◽

Sonia de Lucas Santos

Keyword(s):

Neural Network ◽

Neural Networks ◽

Income Tax ◽

Fraud Detection ◽

Personal Income ◽

Personal Income Tax ◽

Tax Returns ◽

The Neural Networks ◽

Tax Fraud ◽

Efficiency Rate

The goal of the present research is to contribute to the detection of tax fraud concerning personal income tax returns (IRPF, in Spanish) filed in Spain, through the use of Machine Learning advanced predictive tools, by applying Multilayer Perceptron neural network (MLP) models. The possibilities springing from these techniques have been applied to a broad range of personal income return data supplied by the Institute of Fiscal Studies (IEF). The use of the neural networks enabled taxpayer segmentation as well as calculation of the probability concerning an individual taxpayer’s propensity to attempt to evade taxes. The results showed that the selected model has an efficiency rate of 84.3%, implying an improvement in relation to other models utilized in tax fraud detection. The proposal can be generalized to quantify an individual’s propensity to commit fraud with regards to other kinds of taxes. These models will support tax offices to help them arrive at the best decisions regarding action plans to combat tax fraud.

Download Full-text

RLC-GNN: An Improved Deep Architecture for Spatial-Based Graph Neural Network with Application to Fraud Detection

Applied Sciences ◽

10.3390/app11125656 ◽

2021 ◽

Vol 11 (12) ◽

pp. 5656

Author(s):

Yufan Zeng ◽

Jiashan Tang

Keyword(s):

Numerical Experiments ◽

State Of The Art ◽

Single Layer ◽

Fraud Detection ◽

Layer By Layer ◽

Residual Structure ◽

Detection Algorithms ◽

Deep Architecture ◽

Graph Neural Networks ◽

Node Embeddings

Graph neural networks (GNNs) have been very successful at solving fraud detection tasks. The GNN-based detection algorithms learn node embeddings by aggregating neighboring information. Recently, CAmouflage-REsistant GNN (CARE-GNN) is proposed, and this algorithm achieves state-of-the-art results on fraud detection tasks by dealing with relation camouflages and feature camouflages. However, stacking multiple layers in a traditional way defined by hop leads to a rapid performance drop. As the single-layer CARE-GNN cannot extract more information to fix the potential mistakes, the performance heavily relies on the only one layer. In order to avoid the case of single-layer learning, in this paper, we consider a multi-layer architecture which can form a complementary relationship with residual structure. We propose an improved algorithm named Residual Layered CARE-GNN (RLC-GNN). The new algorithm learns layer by layer progressively and corrects mistakes continuously. We choose three metrics—recall, AUC, and F1-score—to evaluate proposed algorithm. Numerical experiments are conducted. We obtain up to 5.66%, 7.72%, and 9.09% improvements in recall, AUC, and F1-score, respectively, on Yelp dataset. Moreover, we also obtain up to 3.66%, 4.27%, and 3.25% improvements in the same three metrics on the Amazon dataset.

Download Full-text