SentiLink has unveiled a new offering aimed at improving financial fraud detection and enhancing proprietary fraud models for financial institutions.
The new tool, Facets, provides feature-specific intelligence derived from SentiLink’s proprietary identity data and its consortium of banks, credit unions and FinTechs, the provider of identity and risk solutions said in a Monday (Sept. 18) press release.
It complements SentiLink’s existing scores targeting synthetic fraud and ID theft, according to the release.
Financial fraud techniques are growing increasingly sophisticated, prompting financial institutions to turn to machine learning (ML)-powered automation to catch fraudsters more efficiently, the release said. However, the effectiveness of these models relies heavily on the quality of the data being used.
With access to more high-quality data, financial institutions can more accurately identify cases of fraud, prevent losses, and approve genuine customers, per the release.
“Many financial institutions today are fighting sophisticated fraudsters with inadequate tools,” SentiLink CEO and Co-founder Naftali Harris said in the release. “With Facets, we’ve unlocked a new level of intelligence that our partners can use to improve and sharpen their models, identify bad actors at account opening, and serve their own customers with the trust they increasingly demand.”
Facets is a powerful enhancement to financial institutions’ fraud models, providing Boolean or numeric attributes that do not require additional feature engineering, according to the release. These attributes are delivered as key-value pairs via the application programming interface (API) response, making them easy to parse. Financial institutions can select the attributes that work best for detecting fraud within their customer population, leading to a significant improvement in accuracy.
Early testing with SentiLink customers has shown that Facets significantly improves the ability to accurately detect fraud, the release said. The solution provides financial institutions with over 250 unique data points from more than 300 institutions and FinTechs that SentiLink collaborates with, enabling better prediction of identity authenticity.
Many of the steps that either prolong or complicate onboarding have to do with banks’ need to comply with know your customer (KYC) and anti-money laundering (AML) regulations, PYMNTS reported in March 2022. Therefore, many banks are prioritizing investments in artificial intelligence (AI)-enabled data analytics features for corporate digital client onboarding.