Wells Fargo Bank, N.a.

Jimmy Wang

Elizabeth Schrag

Michael R Thomas

Jerome Rhodes

Jonathan George Velline

Prasanth Nandanuru

Steve Hatzirallis

Eliza Fain

Dilip Menon

Paramdeep Singh

Ashvil Savio Dcosta

Identification information is received at a transaction machine. The identification information is associated with a user associated with an account. Using an activity profile of the user, a determination is made to enable chained transactions during the user's use of the transaction machine. A sequence of user interfaces to display to the user is generated enabling the user to chain transactions based on the activity profile.

Smart chaining

Examples described herein relate to apparatuses and methods of detecting fraudulent activity at an automated teller machine (ATM) using a machine learning model. A method includes receiving ATM activity data indicative of one or more withdrawal transactions at one or more ATMs using a transaction card, receiving transaction data and ATM data, ingesting the transaction data and the ATM data, analyzing the ingested transaction data and the ingested ATM data using a machine learning model, determining that the ingested transaction data and the ingested ATM data indicate a likelihood of fraudulent activity using the machine learning model, and performing, using the machine learning model, one or more remedial actions based on the determined likelihood of fraudulent activity.

Machine learning based detection of fraudulent acquirer transactions

Identification information is received from a transaction card at a transaction machine. The transaction card is associated with an account holder using the transaction machine. Using an activity profile of the account holder, a determination is made to enable chained transactions during the account holder's use of the transaction machine. A sequence of user interfaces to display to the account holder is generated enabling the account holder to chain transactions based on the account holder chaining transactions during prior transaction machine usage.

Identification information is received from a transaction card at a transaction machine. The transaction card is associated with an account holder using the transaction machine and includes a transaction card type. Using the identification information, activity profile information is accessed for the account holder. The account holder's transaction machine usage is monitored and activity profile information related to the account holder's transaction machine usage is stored. A custom sequence of user interfaces to be displayed to the account holder is generated based on the transaction card type.

Identification information is received from a transaction card at a transaction machine. The transaction card is associated with an account holder using the transaction machine. Using the identification information, activity profile information is accessed for the account holder. The account holder's transaction machine usage is monitored and activity profile information related to the account holder's transaction machine usage is stored. A custom sequence of user interfaces to be displayed to the account holder is generated based on the activity profile information and based on the type of transaction the account holder initiates.

A Check Stack Visualization technique is provided and allows an automated teller machine (ATM) user to see a stack of checks on one screen for check amount confirmation. Customers see all the checks on one screen and indicate that the amounts are correct and click to deposit the checks. In one preferred embodiment, a three dimensional (3D) list is used to display the checks on a Bezier curve along with the amounts highlighted. If the optical character recognition (OCR) process failed on a check amount, the technique highlights the check amount in red. If the customer does not think the check amount is correct, the customer can click on the small check image which brings up a larger check image and they can edit the amount. Once the customer is satisfied that the numbers are correct, the customer can click on a deposit all check button to complete the transaction.

Method and apparatus for check stack visualization

An Any Which Way Check Acceptance method and apparatus is provided, which allow a set of checks to be deposited in a stack, where each check can be oriented in any of sixteen different ways. In one embodiment of the invention, the ATM check reader has an MICR and OCR scanner/reader. When a set of checks is deposited, each check is subjected to a MICR scan and an OCR scan. If the check is aligned correctly and right side up, then the check is processed based on the MICR scan information. If the check is not aligned correctly or is upside down, then the MICR read fails because the magnetic stripe on the check is not where it is expected to be. Rather than rejecting the check at this point, the invention uses the OCR scan information from both the front and back of the check to process the check.

Growing community of inventors

Alamo, CA, United States of America

Jimmy Wang