The Internal Revenue Service has signed a $6.25 million contract with the SAS Institute to use the company’s analytical software to help it identify tax evasion and fraud as part of the agency’s Return Review Program.
The RRP system is being deployed to help the IRS reduce the estimated $345 billion tax gap between taxes owed and paid. The SAS Analytics software will help the IRS improve its fraud detection capabilities and uncover noncompliance at the time a tax return is initially filed.
The IRS is planning to move toward a more real-time tax system in the years ahead that would automatically match electronically filed tax return data with information returns from third parties such as employers and banks. On Thursday, the service held the first of a series of meetings to solicit feedback from accounting and tax practitioner organizations (see
“There is a significant opportunity to avoid improper refunds and payments of large sums of money,” said SAS Federal president Karen Knowles in a statement. “With SAS, the IRS can reduce the number and amount of fraudulent tax refunds, discover emerging fraud schemes and increase tax collections to shrink the tax gap.”
SAS’s analytical software will score tax returns using a combination of business rules, anomaly detection, predictive modeling and social network analysis. Users at the IRS will be able to set up business rules that detect possible fraud and alert investigators and auditors to suspicious-looking returns.
The software searches data for anomalies that could indicate fraud or error. Predictive modeling techniques employ historical behavioral information to identify suspect behavior similar to known fraud patterns. Social network analysis uncovers hidden relationships or linkages that suggest collusion and organized fraud rings.
Another tool to combat fraud, SAS Text Miner, will scour unstructured data, such as from call centers, to detect suspicious activity. Alerts and results will be reported to the IRS via a customizable dashboard. Case management capabilities in the software will also help IRS investigators prioritize and assign cases.
The $6.25 million contract will give the IRS Return Review Program access to many SAS technologies, including SAS Fraud Framework for Government, SAS Social Network Analysis and SAS's data integration, data mining and business intelligence technologies. To analyze the massive amounts of data in the system, the IRS will also be able to leverage SAS In-Database, SAS Scoring Accelerator for Greenplum and SAS Grid Manager.
SAS’s software is also used by other government agencies to detect Medicare and Medicaid fraud, purchase-card fraud, bid-rigging and terrorist financing.
