Fraudulent Act Detection Task
A Fraudulent Act Detection Task is an anomaly detection task that requires the detection of fraudulent acts.
- AKA: Fraud Recognition.
- Context:
- measure: Fraud Incident Count, Fraud-Loss Measure, ...
- ...
- It can (typically) have a Benefit-Cost Ratio that quickly reduces with time.
- ...
- It can range from being a Human-Performed Fraud Detection Task to being a Automated Fraud Detection Task.
- It can range from being a Real-Time Fraud Detection Task to being a Batch Fraud Detection Task, based on the urgency of detection and processing requirements.
- ...
- It can be supported by a Fraud Detection System (that implements a fraud detection algorithm).
- …
- Example(s):
- Financial Fraud Detection:
- Credit Card Fraud Detection (to detect unauthorized credit card use).
- Insurance Fraud Detection (to identify false insurance claims).
- Digital Fraud Detection:
- Web Fraud Detection (to identify suspicious online activities).
- Identity Theft Detection (to uncover identity impersonation in digital platforms).
- Regulatory Fraud Detection:
- Tax Fraud Detection (to identify evasion of tax liabilities).
- Compliance Fraud Detection (to detect violations of regulatory requirements).
- …
- Financial Fraud Detection:
- Counter-Example(s):
- Hoax Detection (of hoaxes).
- Spam Detection (of spams).
- See: Deceptive Behavior Detection, Predictive Analytics, Deception, Predictive Analytics, Financial Transaction, Identity Theft.
References
2020
- (Wikipedia, 2020) ⇒ https://en.wikipedia.org/wiki/Fraud#Detection Retrieved:2020-10-21.
- The detection of fraudulent activities on a large scale is possible with the harvesting of massive amounts of financial data paired with predictive analytics or forensic analytics, the use of electronic data to reconstruct or detect financial fraud.
Using computer-based analytic methods in particular allows for surfacing of errors, anomalies, inefficiencies, irregularities, and biases which often refer to fraudsters gravitating to certain dollar amounts to get past internal control thresholds.[1] These high-level tests include tests related to Benford's Law and possibly also those statistics known as descriptive statistics. High-level tests are always followed by more focused tests to look for small samples of highly irregular transactions. The familiar methods of correlation and time-series analysis can also be used to detect fraud and other irregularities.
- The detection of fraudulent activities on a large scale is possible with the harvesting of massive amounts of financial data paired with predictive analytics or forensic analytics, the use of electronic data to reconstruct or detect financial fraud.
2016a
- (Wikipedia, 2016) ⇒ http://wikipedia.org/wiki/predictive_analytics#Fraud_detection Retrieved:2016-3-8.
- Fraud is a big problem for many businesses and can be of various types: inaccurate credit applications, fraudulent transactions (both offline and online), identity thefts and false insurance claims. These problems plague firms of all sizes in many industries. Some examples of likely victims are credit card issuers, insurance companies, retail merchants, manufacturers, business-to-business suppliers and even services providers. A predictive model can help weed out the "bads" and reduce a business's exposure to fraud.
Predictive modeling can also be used to identify high-risk fraud candidates in business or the public sector. Mark Nigrini developed a risk-scoring method to identify audit targets. He describes the use of this approach to detect fraud in the franchisee sales reports of an international fast-food chain. Each location is scored using 10 predictors. The 10 scores are then weighted to give one final overall risk score for each location. The same scoring approach was also used to identify high-risk check kiting accounts, potentially fraudulent travel agents, and questionable vendors. A reasonably complex model was used to identify fraudulent monthly reports submitted by divisional controllers.
The Internal Revenue Service (IRS) of the United States also uses predictive analytics to mine tax returns and identify tax fraud.
Recentadvancements in technology have also introduced predictive behavior analysis for web fraud detection. This type of solution utilizes heuristics in order to study normal web user behavior and detect anomalies indicating fraud attempts.
- Fraud is a big problem for many businesses and can be of various types: inaccurate credit applications, fraudulent transactions (both offline and online), identity thefts and false insurance claims. These problems plague firms of all sizes in many industries. Some examples of likely victims are credit card issuers, insurance companies, retail merchants, manufacturers, business-to-business suppliers and even services providers. A predictive model can help weed out the "bads" and reduce a business's exposure to fraud.
2016b
- (Wikipedia, 2016) ⇒ http://wikipedia.org/wiki/fraud#Detection Retrieved:2016-3-8.
- For detection of fraudulent activities on the large scale, massive use of (online) data analysis is required, in particular predictive analytics or forensic analytics. Forensic analytics is the use of electronic data to reconstruct or detect financial fraud. The steps in the process are data collection, data preparation, data analysis, and the preparation of a report and possibly a presentation of the results. Using computer-based analytic methods Nigrini's wider goal is the detection of fraud, errors, anomalies, inefficiencies, and biases which refer to people gravitating to certain dollar amounts to get past internal control thresholds. The analytic tests usually start with high-level data overview tests to spot highly significant irregularities. In a recent purchasing card application these tests identified a purchasing card transaction for 3,000,000 Costa Rica Colons. This was neither a fraud nor an error, but it was a highly unusual amount for a purchasing card transaction. These high-level tests include tests related to Benford's Law and possibly also those statistics known as descriptive statistics. These high-tests are always followed by more focused tests to look for small samples of highly irregular transactions. The familiar methods of correlation and time-series analysis can also be used to detect fraud and other irregularities. Forensic analytics also includes the use of a fraud risk-scoring model to identify high risk forensic units (customers, employees, locations, insurance claims and so on). Forensic analytics also includes suggested tests to identify financial statement irregularities, but the general rule is that analytic methods alone are not too successful at detecting financial statement fraud.
- (Wikipedia, 2016) ⇒ http://wikipedia.org/wiki/Fraud#Detection Retrieved:2016-3-8.
- In law, fraud is deliberate deception to secure unfair or unlawful gain, or to deprive a victim of a legal right. Fraud itself can be a civil wrong (i.e., a fraud victim may sue the fraud perpetrator to avoid the fraud and/or recover monetary compensation), a criminal wrong (i.e., a fraud perpetrator may be prosecuted and imprisoned by governmental authorities) or it may cause no loss of money, property or legal right but still be an element of another civil or criminal wrong. The purpose of fraud may be monetary gain or other benefits, such as obtaining a driver's license or qualifying for a mortgage by way of false statements.
A hoax is a distinct concept that involves deliberate deception without the intention of gain or of materially damaging or depriving a victim.
- In law, fraud is deliberate deception to secure unfair or unlawful gain, or to deprive a victim of a legal right. Fraud itself can be a civil wrong (i.e., a fraud victim may sue the fraud perpetrator to avoid the fraud and/or recover monetary compensation), a criminal wrong (i.e., a fraud perpetrator may be prosecuted and imprisoned by governmental authorities) or it may cause no loss of money, property or legal right but still be an element of another civil or criminal wrong. The purpose of fraud may be monetary gain or other benefits, such as obtaining a driver's license or qualifying for a mortgage by way of false statements.
2009
- (Chandola et al., 2009) ⇒ Varun Chandola, Arindam Banerjee, and Vipin Kumar. (2009). “Anomaly Detection: A survey.” In: ACM Computing Surveys, 41(3) doi:10.1145/1541880.1541882
2004
- (Hodge & Austin, 2004) ⇒ Victoria Hodge, and Jim Austin. (2004). “A Survey of Outlier Detection Methodologies.” In: Journal Artificial Intelligence Review, 22(2). doi:10.1023/B:AIRE.0000045502.10941.a9
- QUOTE: Fraud detection - detecting fraudulent applications for credit cards, state benefits or detecting fraudulent usage of credit cards or mobile phones.
2002
- (Bolton & Hand, 2002) ⇒ Richard J. Bolton, and David J. Hand. (2002). “Statistical Fraud Detection: A review.” In: Statistical Science, 17(3).