Artificial Intelligence | News | Insights | AiThority
[bsfp-cryptocurrency style=”widget-18″ align=”marquee” columns=”6″ coins=”selected” coins-count=”6″ coins-selected=”BTC,ETH,XRP,LTC,EOS,ADA,XLM,NEO,LTC,EOS,XEM,DASH,USDT,BNB,QTUM,XVG,ONT,ZEC,STEEM” currency=”USD” title=”Cryptocurrency Widget” show_title=”0″ icon=”” scheme=”light” bs-show-desktop=”1″ bs-show-tablet=”1″ bs-show-phone=”1″ custom-css-class=”” custom-id=”” css=”.vc_custom_1523079266073{margin-bottom: 0px !important;padding-top: 0px !important;padding-bottom: 0px !important;}”]

Point Predictive Awarded Third Patent for Enhanced Risk Detection

Newly awarded patent helps lenders better identify the risk profiles of their dealer relationships

Point Predictive Inc., the San Diego-based company that provides machine learning solutions to lenders, announced that the U.S. Patent Office has granted the company another patent for its powerful machine learning technology for detection of risky behavior patterns of car dealerships. The patent, the third issued to Point Predictive this year, covers the company’s method for assessing dealer risk by scoring loan application submissions in real-time across multiple lenders and then aggregating that risk at the dealer level. The technology leverages the concept that the changing risk of a dealer is directly correlated to the changing risk level in the deals that dealer submits to a lender over time.

“We’re excited to receive this third patent,” said Tim Grace, CEO of Point Predictive. “Our mission has always been to power trust in lending with our AI and this patent strengthens our solution portfolio. For a lender, understanding the risk of the dealer submitting an auto loan application is often as important as understanding the risk of the borrower. If a dealer shows a sudden high-risk pattern, lenders want to be alerted so they can take preventive measures in their application decisioning process – not after that fact when loans fund and later default. Conversely, if a dealership’s profile shows lowered risk over time, lenders may want to improve the pull-through from those dealerships by providing enhanced dealer incentives.”

Recommended AI News: Epicor Announces the Release of DocStar ECM for ArcGIS

U.S. Patent No. 10,733,668, “Multi-Layer Machine Learning Classifier,” was granted to Point Predictive on August 4, 2020. The underlying technology is currently used in two solutions: Auto Fraud Manager and DealerTrace. This third patent completes the initial intellectual property protection coverage relating to the company’s machine learning technology for auto lending fraud detection.

Auto Fraud Manager helps lenders identify misrepresentation that could lead to default on high-risk applications, while streamlining the approval of low-risk applications. Lenders have found that streamlining low-risk applications by clearing unnecessary stipulations not only results in a smoother funding process, but also helps increase pull through of approved applications by up to 50%.

Recommended AI News: MRM Names Jayna Kothary As Global Chief Technology Officer

DealerTrace leverages the fraud and early payment default detection analytics of Auto Fraud Manager to create dealer-centric risk assessments based on lender-reported application and outcome data. Fuzzy-matching techniques are used to recognize the same dealer – perhaps operating under different names or with slightly different demographic characteristics – across multiple participating lenders. This allows DealerTrace to form a cross-lender consortium-level view of each dealer’s risk characteristics beyond what any individual lender could do on its own. It also enables participating lenders to benchmark their book of loans from a particular dealer against the consortium’s view of that dealer. This is especially useful for detecting instances of “adverse selection” where a dealer may be routing its riskier loans to certain lenders.

Recommended AI News: Nissan Moves to Oracle Cloud Infrastructure for High-Performance Computing

Leave A Reply

Your email address will not be published.