Pointpredictive, Inc.

Corelogic Solutions, LLC

Frank J McKenna

Gregory Gancarz

Timothy J Grace

Michael J Kennedy

Jacqueline Marie Doty

Susan Harper Allen

The present disclosure relates generally to a risk-based fraud identification and risk analysis system. For example, the system may receive application data from a first borrower user, determine a segment associated with the application data, apply application data to one or more machine learning (ML) models, and receive a score based at least in part upon output of the ML model.

Risk-based machine learning classifier

The present disclosure relates generally to a multi-layer fraud identification and risk analysis system. For example, the system may receive application data from a first borrower user, apply application data to one or more machine learning (ML) models, and receive a first score based at least in part upon output of the ML model that is associated with the first borrower user. The system may aggregate scores associated with multiple borrower users to a cumulative dealer user level. The aggregated first scores associated with the dealer user, as well as other corresponding application data, may be provided as input to a second ML model. Output from the second ML model may be associated with the dealer user as a second score.

Multi-layer machine learning classifier

The present disclosure relates generally to a multi-layer fraud identification and risk analysis system. For example, the system may receive a plurality of first scores associated with borrower users and a dealer user based at least in part upon output of the first ML model. The system may receive a request from a lender user device for a second score, where the dealer user and the lender user device are associated according to a correlative score. The plurality of applications and the correlative score may be used as input to the second ML model that quantifies the risk of the dealer user specifically to the lender user, based on attributes associated with the application data, dealer user, and/or lender user. Output from the second ML model may be provided to the lender user device.

Multi-layer machine learning classifier with correlative score

Risk-based machine learning classsifier

A system is disclosed for identifying potential short sale transactions that can be bundled or combined to produce a beneficial result to the financial institutions involved. Embodiments are directed to the identification of properties matching certain criteria that make them likely candidates for a short sale transaction. Other embodiments disclosed herein assess the value of identified properties and associated mortgages to facilitate short sale transactions, so that mortgage holders can have an accurate assessment of potential losses involved in short sale and foreclosure situations where multiple mortgages are involved. One embodiment provides an assessment of estimated potential losses realized by secondary mortgage holders and/or losses avoided by primary mortgage holders in proposed short sale situations. The assessment enables financial institutions involved in certain short sales to offset losses realized and/or avoided with those financial institutions that may have opposite mortgage positions (primary vs. secondary) on other properties.

Growing community of inventors

La Jolla, CA, United States of America

Frank J McKenna