A novel method for detecting credit card fraud problems.

Credit card fraud is a significant problem that costs billions of dollars annually. Detecting fraudulent transactions is challenging due to the imbalance in class distribution, where the majority of transactions are legitimate. While pre-processing techniques such as oversampling of minority classes are commonly used to address this issue, they often generate unrealistic or overgeneralized samples. This paper proposes a method called autoencoder with probabilistic xgboost based on SMOTE and CGAN(AE-XGB-SMOTE-CGAN) for detecting credit card frauds.AE-XGB-SMOTE-CGAN is a novel method proposed for credit card fraud detection problems. The credit card fraud dataset comes from a real dataset anonymized by a bank and is highly imbalanced, with normal data far greater than fraud data. Autoencoder (AE) is used to extract relevant features from the dataset, enhancing the ability of feature representation learning, and are then fed into xgboost for classification according to the threshold. Additionally, in this study, we propose a novel approach that hybridizes Generative Adversarial Network (GAN) and Synthetic Minority Over-Sampling Technique (SMOTE) to tackle class imbalance problems. Our two-phase oversampling approach involves knowledge transfer and leverages the synergies of SMOTE and GAN. Specifically, GAN transforms the unrealistic or overgeneralized samples generated by SMOTE into realistic data distributions where there is not enough minority class data available for GAN to process effectively on its own. SMOTE is used to address class imbalance issues and CGAN is used to generate new, realistic data to supplement the original dataset. The AE-XGB-SMOTE-CGAN algorithm is also compared to other commonly used machine learning algorithms, such as KNN and Light GBM, and shows an overall improvement of 2% in terms of the ACC index compared to these algorithms. The AE-XGB-SMOTE-CGAN algorithm also outperforms KNN in terms of the MCC index by 30% when the threshold is set to 0.35. This indicates that the AE-XGB-SMOTE-CGAN algorithm has higher accuracy, true positive rate, true negative rate, and Matthew's correlation coefficient, making it a promising method for detecting credit card fraud.

Advanced detection tools in food fraud: A systematic review for holistic and rational detection method based on research and patents.

Modern food chain supply management necessitates the dire need for mitigating food fraud and adulterations. This holistic review addresses different advanced detection technologies coupled with chemometrics to identify various types of adulterated foods. The data on research, patent and systematic review analyses (2018-2023) revealed both destructive and non-destructive methods to demarcate a rational approach for food fraud detection in various countries. These intricate hygiene standards and AI-based technology are also summarized for further prospective research. Chemometrics or AI-based techniques for extensive food fraud detection are demanded. A systematic assessment reveals that various methods to detect food fraud involving multiple substances need to be simple, expeditious, precise, cost-effective, eco-friendly and non-intrusive. The scrutiny resulted in 39 relevant experimental data sets answering key questions. However, additional research is necessitated for an affirmative conclusion in food fraud detection system with modern AI and machine learning approaches.

The court's judgment on the true identity of the author responsible for identity fraud and the consequent batch retraction of Iranian's papers in the journal of diagnostic pathology.

This letter concerns retracted papers published in the Journal of Diagnostic Pathology, where my name was misused as the author or corresponding author without my permission or knowledge. Considering that all misconducts were directed by an author during initial manuscripts' submissions, I opened a case in Iran's Cyber Police (FATA) to unravel the true identity of the submitting author. After Cyber Police's report revealed the true identity of the submitting author, the court started a thorough investigation and finally convicted the submitting author for identity fraud and data forgery through creating and using fake email addresses.

Anticounterfeiting and Fraud Mitigation Solutions for High-value Food Products.

Globalization and the increasing complexity of supply chains have allowed food fraud to expand to a great extent. Some of the most serious effects of these deceitful activities are damage to a brand's reputation and trust, economic losses, and public health risks. The usual victims of food fraud are dairy, meat, fish, and seafood products, as well as fats/oils and alcoholic drinks. The purpose of this review paper is to present an updated analysis of the currently available anticounterfeit technologies and their application to the four most fraud-affected food supply chains. An assessment that was conducted to determine when the adoption of a combination of technologies could enhance food safety and brand protection is also provided. The obtained results indicate that electronic and data-driven technologies (RFID devices and digital traceability systems) are still in their infancy in the food sectors that are subjected the most to fraudulent activities. Research is necessary to develop innovative digital and physical technologies to "outsmart" such fraudsters and to prevent their illicit actions in the food sector.

Research on the risk governance of fraudulent reimbursement of patient consultation fees.

Background: The fundamental medical insurance fund, often referred to as the public's "life-saving fund," plays a crucial role in both individual well-being and the pursuit of social justice. Medicare fraudulent claims reduce "life-saving money" to "Tang's monk meat", undermining social justice and affecting social stability. Methods: We utilized crawler technology to gather textual data from 215 cases involving fraudulent health insurance claims. Simultaneously, statistical data spanning 2018 to 2021 was collected from the official websites of the China Medical Insurance Bureau and Anhui Medical Insurance Bureau. The collected data underwent comprehensive analysis through Excel, SPSS 26.0 and R4.2.1. Differential Auto-Regressive Moving Average Model (ARIMA (p, d, q)) was used to fit the fund safety forecast model, and test the predictive validity of the forecast model on the fund security data from July 2021 to October 2023 (the fund security data of Anhui Province from September 2021 to October 2023). Results: The outcomes revealed that fraudulent claims by health insurance stakeholders adversely impact the equity of health insurance funds. Furthermore, the risk management practices of Medicare fund administrators influence the publication of fraudulent claims cases. Notably, differences among Medicare stakeholders were observed in the prevalence of fraudulent claims. Additionally, effective governance of fraudulent claims risks was found to have a positive impact on the overall health of healthcare funds. Moreover, the predictive validity of the forecast model on the national and Anhui province's fund security data was 92.86% and 100% respectively. Conclusion: We propose four recommendations for the governance of health insurance fraudulent claims risk behaviors. These recommendations include strategies such as "combatting health insurance fraudulent claims to preserve the fairness of health insurance funds", "introducing initiatives for fraud risk governance and strengthening awareness of the rule of law", "focusing on designated medical institutions and establishing a robust long-term regulatory system", and "adapting to contemporary needs while maintaining a focus on long-term regulation".

Radiology residency oversight: A Qui Tam wake-up call.

Like every physician practice, academic radiology practices must pay heed to all governmental regulations. The federal False Claims Act serves to protect US taxpayers and requires strict adherence. Violations, often brought forth by whistleblowers, can carry steep financial repercussions.

The Assessment of Situational Judgement Questionnaire: A Novel Instrument to Detect Susceptibility to Financial Scamming.

BACKGROUND: Existing measures of scam susceptibility lack ecological validity and situational variability. Evidence suggests that all adults may be susceptible to scams, though a comprehensive fraud victimization theory remains to be explored. OBJECTIVE: To identify cognitive and sociodemographic variables that differentiate individuals with high scam susceptibility from those less susceptible. This article describes the development and feasibility of the Assessment of Situational Judgment questionnaire (ASJ), a brief tool designed to detect scam susceptibility. METHODS: The 17-item ASJ was developed using a combination of existing scams reported by the Florida Division of Consumer Services and legitimate scenarios. Participants were presented with scam and legitimate scenarios and queried regarding their willingness to engage. Response options were offered with instructions on a 7-point Likert scale (extremely unlikely to extremely likely). Pilot data from a development sample provided the foundation for the final version of the ASJ. RESULTS: The final version of the ASJ was administered to 183 online participants. The Scam factor (8 items) explained 50.6% of the variance. The Legit factor (9 items) reported on a 7-point Likert scale explaining 10.6% of the variance. A Scam to Legit ratio provides a proxy for overall scam susceptibility. Cut-off scores of 24 on the Scam factor, 47 on the Legit factor, and 0.62 on the ratio optimize measures of scam susceptibility. CONCLUSIONS: The ASJ is a brief, ecologically valid measure of scam susceptibility. There is a need for a sensitive and specific tool to detect scam susceptibility in clinical, community, and financial settings.

Database of Food Fraud Records: Summary of Data from 1980 to 2022.

Food fraud prevention and detection remains a challenging problem, despite recent developments in regulatory and auditing requirements. In 2012, the United States Pharmacopeial Convention created a database of food ingredient fraud. The objective of this research was to report on updates made to the database structure and to provide an updated analysis of food fraud records. The restructured database was relational and included four tables: ingredients, adulterants, adulteration records, and references. Four adulteration record types were created to capture the variety of information that can be found in public food fraud reports. Information was searched and extracted from the peer-reviewed scientific literature, media publications, regulatory reports, judicial records, trade association reports, and other public sources covering 1980-present. Over an almost seven-year data entry period, a total of 15,575 records were entered, sourced primarily from the peer-reviewed literature and media reports. The percentage of records that included at least one potentially hazardous adulterant ranged from 34% to 60%, depending on the record type. The ingredients with the highest number of incident and inference records included fluid cow's milk, extra virgin olive oil, honey, beef, and chili powder. The ingredient groups with the highest number of incident and inference records included Dairy Ingredients, Seafood Products, Meat and Poultry Products, Herbs, Spices, and Seasonings, Milk and Cream, and Alcoholic Beverages. This database was created to serve as a standardized source of information about publicly documented occurrences of food fraud and other information relevant to fraud risk to support food fraud vulnerability assessments, mitigation plans, and food safety plans. These data support the contention that food fraud presents a public health risk that should continue to be addressed by food safety systems worldwide.

What are people's attitudes toward medical insurance fraud whistleblowers? a study from China.

PURPOSE: Medical insurance fraud has caused huge losses to countries around the world, and public reporting has become an important means to combat medical insurance fraud. The attitude of medical insurance fraud whistleblowers affects people's reporting behavior, and understanding people's attitude toward medical insurance fraud whistleblowers provides a basis for further improving the system and policy of public participation in medical insurance fund supervision. METHODS: We adopted the questionnaire method to conduct a national cross-sectional survey of the Chinese public and analyzed the data using Chi-square tests, Fisher's exact tests, and binary logistic regression models. RESULTS: A total of 837 respondents were included, and 81.8% of the population had a supportive attitude toward medical insurance fraud whistleblowers, with gender, whether they had used medical insurance reimbursement, and present life satisfaction being statistically significant (P < 0.05). CONCLUSION: The public is generally supportive of medical insurance fraud whistleblowers, and women, those who have used medical insurance for reimbursement, and those who are satisfied with their lives are more likely to be supportive of medical insurance fraud whistleblowers.

What influences the public's willingness to report health insurance fraud in familiar or unfamiliar healthcare settings? a cross-sectional study of the young and middle-aged people in China.

INTRODUCTION: Young and middle-aged people are important participants in the fight against health insurance fraud. The study aims to investigate the differences in their willingness to report health insurance fraud and the factors influencing it when it occurs in familiar or unfamiliar healthcare settings. METHODS: Data were obtained from a validated questionnaire from 828 young and middle-aged people. McNemar's test was used to compare the public's willingness to report under the two scenarios. Chi-square tests and multiple logistic regression analysis were used to analyze the determinants of individuals' willingness to report health insurance fraud in different scenarios. RESULTS: Young and middle-aged people were more likely to report health insurance fraud in a familiar healthcare setting than in an unfamiliar one (McNemar's χ²=26.51, P < 0.05). Their sense of responsibility for maintaining the security of the health insurance fund, the government's openness about fraud cases, and the perception of their ability to report had significant positive effects on the public's willingness to report in both settings (P < 0.05). In a familiar healthcare setting, the more satisfied the public is with government measures to protect whistleblowers, the more likely they are to report (OR = 1.44, P = 0.025). Those who perceive the consequences of health insurance fraud to be serious are more likely to report than those who perceive the consequences to be less serious (OR = 1.61, P = 0.042). CONCLUSION: Individuals are more likely to report health insurance fraud in familiar healthcare settings than in unfamiliar ones, in which their awareness of the severity of the consequences of health insurance fraud and their perceived risk after reporting it play an important role. The government's publicizing of fraud cases and enhancing the public's sense of responsibility and ability to maintain the safety of the health insurance fund may be a way to increase their willingness to report, regardless of whether they are familiar with the healthcare setting or not.

Case Studies of Fraud Associated with the Use of Radiation Sources: Practical Avoidance Strategies Based on Lessons Learned.

ABSTRACT: Periodically the radiation protection profession has experienced purposeful deception practices that remained undetected for some time. Upon discovery, the cases of fraud revealed gaps in confirmation or validation practices that the radiation protection community should note. Summarized here is a convenience sample of actual cases of fraud involving radiation sources along with the exploited process vulnerabilities. Recommended process improvements that the radiation safety community may consider are presented to improve the collective fidelity of radiation protection processes.

Beyond the individual: An improved telecom fraud detection approach based on latent synergy graph learning.

The development of telecom technology not only facilitates social interactions but also inevitably provides the breeding ground for telecom fraud crimes. However, telecom fraud detection is a challenging task as fraudsters tend to commit co-fraud and disguise themselves within the mass of benign ones. Previous approaches work by unearthing differences in calling sequential patterns between independent fraudsters, but they may ignore synergic fraud patterns and oversimplify fraudulent behaviors. Fortunately, graph-like data formed by traceable telecom interaction provides opportunities for graph neural network (GNN)-based telecom fraud detection methods. Therefore, we develop a latent synergy graph (LSG) learning-based telecom fraud detector, named LSG-FD, to model both sequential and interactive fraudulent behaviors. Specifically, LSG-FD introduces (1) a multi-view LSG extractor to reconstruct synergy relationship-oriented graphs from the meta-interaction graph based on second-order proximity assumption; (2) an LSTM-based calling behavior encoder to capture the sequential patterns from the perspective of local individuals; (3) a dual-channel based graph learning module to alleviate the disassortativity issue (caused by the camouflages of fraudsters) by incorporating the dual-channel frequency filters and the learnable controller to adaptively aggregate high- and low-frequency information from their neighbors; (4) an imbalance-resistant model trainer to remedy the graph imbalance issue by developing a label-aware sampler. Experiment results on the telecom fraud dataset and another two widely used fraud datasets have verified the effectiveness of our model.

Federal Enforcement of Pharmaceutical Fraud under the False Claims Act, 2006-2022.

CONTEXT: The False Claims Act is the US federal government's primary tool for identifying and penalizing pharmaceutical fraud. The Department of Justice uses the False Claims Act to bring civil cases against drug manufacturers that allegedly obtain improper payment from federal programs. METHODS: The authors searched the Department of Justice website for press releases published between 2006 and 2022 that announced fraud actions brought against drug companies. They then used the World Health Organization's Anatomical Therapeutic Classification index to identify the classes of prescription drugs implicated in fraud actions. FINDINGS: During fiscal years 2006-2022, payments by six manufacturers amounted to more than 28% of total payments made as a result of federal False Claims Act actions. Nervous system and cardiovascular drugs were the classes of medications most commonly implicated in alleged fraud. Federal officials most frequently alleged that companies improperly promoted nervous system drugs and paid kickbacks to increase revenues from cardiovascular, antineoplastic and immunomodulating, and alimentary tract and metabolism drugs. CONCLUSIONS: Despite frequent pharmaceutical fraud settlements and penalties, incidence of alleged fraud among drug companies remains high. Alternative methods for preventing and deterring fraud could help safeguard our health systems and promote public health, and policy makers should ensure that effective fraud enforcement complements preventive public health regulation.

Features in visible and Fourier transform infrared spectra confronting aspects of meat quality and fraud.

Rapid assessment of microbiological quality (i.e., Total Aerobic Counts, TAC) and authentication (i.e., fresh vs frozen/thawed) of meat was investigated using spectroscopic-based methods. Data were collected throughout storage experiments from different conditions. In total 526 spectra (Fourier transform infrared, FTIR) and 534 multispectral images (MSI) were acquired. Partial Least Squares (PLS) was applied to select/transform the variables. In the case of FTIR data 30 % of the initial features were used, while for MSI-based models all features were employed. Subsequently, Support Vector Machines (SVM) regression/classification models were developed and evaluated. The performance of the models was evaluated based on the external validation set. In both cases MSI-based models (Root Mean Square Error, RMSE: 0.48-1.08, Accuracy: 91-97 %) were slightly better compared to FTIR (RMSE: 0.83-1.31, Accuracy: 88-94 %). The most informative features of FTIR for the case of quality were mainly in 900-1700 cm-1, while for fraud the features were more dispersed.

Living-skin Detection using Multi-layer Skin Property Perceived by the Structured Light.

Living-skin detection has been used to prevent the attack of face fraud in a face recognition system. In this paper, we propose a new concept that exploits the multi-layer structure property of skin for living-skin detection. We observe a significant difference in the blur of the laser spot created by the structured light on the skin and non-skin due to the characteristic properties of laser photons in skin penetration and reflection. Based on this observation, we designed a new living-skin detection algorithm to differentiate skin and non-skin based on the blur detection of laser spots. The experimental results show that the proposed setup and method have a promising performance with an averaged precision of 96.7%, averaged recall of 82.2%, and averaged F1-score of 88.6% on a dataset of 20 adult subjects. This demonstrates the effectiveness of the new concept that uses multi-layer properties of skin tissues for living-skin detection, which may lead to new solutions for face anti-spoofing.