The ARGO Fraud solution, OASIS™ (Optimized Assessment of Suspicious Items), provides cross-channel, multi-fund analytics and adjudication workflow to detect fraudulent transactions and suspicious items. ARGO’s innovative technology and advanced analytics enable more than 500 financial institutions in the U.S. and Canada to increase fraud prevention and detection beyond the capabilities of legacy systems. Fraud detection improves by combining software innovations such as decision tree/multiple variable analysis, image analysis, and machine learning predictive analytics. Data reference topology increases to include contextual information and negative historical analytics. In turn, these outcomes detect transactional fraud and suspicious activity, reduce false negatives, and enable a financial institution to make better fraud-related decisions faster. “The OASIS solution provides a centralized interface for suspect adjudication workflow and enables case management for those items requiring additional research,” says David Engebos, ARGO’s President and COO.
OASIS uses extensive mathematical and machine learning methods that are defined in the context of configurable parameters. Machine learning, especially adept at pattern recognition in large data sets, is useful for determining ‘good’ versus ‘bad.’ Mathematical probability then measures degrees of good and bad and assigns a relative numerical probability value on a suspicious transaction. This happens by analyzing both images and transactional data and identifying potential suspicious items and corresponding risk exposure.
To better explain the efficacy of OASIS, Engebos shares the case study of a regional bank. The financial institution was using a combination of internal controls and prior generation fraud software to identify and stop fraud attempts. This process generated a large number of false positives, and they had to spend a great deal of time reviewing non-fraud transactions that were flagged as suspect. This bank installed OASIS to perform both transaction and image analysis. As a result, they could significantly improve detection accuracy based on decisioning inputs and scoring made by machine-learning models. This played a vital role in reducing false negatives, financial losses, operating expenses, and reputational risk.
ARGO uses a Champion-Challenger mechanism to deploy new models, track and compare results, and then recommend when and how to switch between the current Champion model and the Challenger models. While Champion is the model that produces the official fraud model risk score, Challenger executes alongside the Champion. “Scores are recorded but are not used by OASIS.
The OASIS solution provides a centralized interface for suspect adjudication workflow and enables case management for those items requiring additional research
Further, OASIS offers complete Regulation CC hold functionality in real-time at the teller line with fully-automated recommended hold amounts based on defined thresholds and durations. The solution automates the work of determining which items a hold should be placed on, how much the hold should be, and then automatically places the hold, minimizing Reg CC decision-making by the teller. OASIS also supports case-by-case and exception hold processing. In both cases, the solution notifies the teller of the available dates for each hold and applicable hold reasons and then prints the hold notice for the customer.
To address fraud related to cash transactions and threats of money laundering, OASIS AML identifies potential money laundering and terrorist financing activities and enforces due diligence. The solution’s comprehensive case management capabilities and integrated workflow help financial institutions reduce regulatory risks, reputational risks, and financial losses. OASIS AML also enables compliance officers to learn more about customers by providing automated risk analysis at account opening and throughout the customer relationship. The solution provides enhanced due diligence tools to create customer-risk profiles based on customer information, watch-list searches, and analysis of key risk indicators. It also mitigates risks and identifies high-risk customers through ongoing analysis, and maintains and updates client information for continuing due diligence. “Using customer information, files are analyzed based on attributes, such as products and services utilized, type of business, or income and address demographics,” explains Engebos.
ARGO’s increased accuracy in fraud detection has a direct impact in all cost components of running fraud prevention departments, including deposits, retail, IT, and certainly impacting the bottom-line. “We continue to significantly invest in several areas, including constantly improving analytical capabilities and expanding on the detection and prevention features,” says Engebos. Investment in R&D will increase the ability to detect more fraud and fraud types accurately while reducing false positives, cost, customer friction, and dissatisfaction.