Bank of America Corporation

Xianhong Zhang

Kapil Pruthi

Andrew T Keys

Michael E Toth

Daniel Lynn Carpenter

Mark A Pender

Apeksh M Dave

Elizabeth Votaw

Tao Huang

Hitesh Shah

Ashish Arora

Raghav Shenoy

Muniraju Jayaramaiah

Arul Arasu Madavadiyan

Palash K Shil

Andy Keys

Vikram Jalota

Wenhui Meng

Kalyan V Pasumarthi

Archie Agrawal

Dennis M Osborne

Srinivasa Rao Goriparthi

Ramakrishna R Yannam

Yu Fu

Spencer Yezo

Heather Roseann Dolan

Nishant H Shah

David Charles Shroyer

Kent Steger

Gail R Davis

Kris Deion Tullos

Jessica Lynn Brandt

Aaron Daniel Lewis

Karen Choy

Jeffrey Jacoby

William Larry Grant, Jr

Linda Marie Dziedzic

Duncan Converse Morrell

Justin McCord

Ronald C Mitchell, Jr

Michael Dean Bridges

Craig T Johnson

Tanuja Melkani

Debra Jean Lewis

Joseph M Hollmann

Akshay Jain

Sidney A Winchester, Jr

WenTsong Shiue

Sateesh Siripurapu

Manoj Shankar

Sridevi Rachakulla

A telephonic routing system to enable rerouting a call between a live agent and an automated agent. The system may include a receiver configured to receive a telephonic call from a caller. The system may also include a router configured to route the telephonic call to a first communication pathway where the first communication pathway may connect the telephonic call to a live agent headset device. The system may also include a processor configured to retrieve a customer profile associated with a phone number of the telephonic call and display authentication data, on a graphical user interface ('GUI') of a computing device associated with the live agent headset device. The router may be further configured to reroute the telephonic call to a second communication pathway for verifying an authenticity of the caller. The second communication pathway may establish a connection between the telephonic call and the automated agent.

Enhanced interactive voice response system providing bi-communication pathways for call support and routing

A one-way data migration system may shift authentication data from a legacy database to a current database. The system may include one or more databases, a mobile device, a mobile device processor, a backend server, a backend receiver, a backend processor, and a backend transmitter. The backend transmitter may transmit requests to the mobile device. The backend receiver may receive the usernames and passwords. The backend processor may authenticate the input usernames and passwords using a hash previously stored within a first database. The backend processor may create a current or updated hash for the password. The system may change the underlying pathway from a pathway associated with the hash previously stored within the first database to a current pathway associated with the current hash. A second database may store the current hash. Following storage of the current hash, the processor may flag the username and/or password as migrated.

One way data migration system

Aspects of the disclosure relate to processing authentication requests to secured information systems based on machine-learned user behavior profiles. A computing platform may receive an authentication request corresponding to a request for a user of a client computing device to access one or more secured information resources associated with a user account. The computing platform may capture behavioral parameters associated with the client computing device and may evaluate the behavioral parameters using a behavioral profile associated with the user account to determine a behavioral deviation score. Based on the behavioral deviation score, the computing platform may select an authentication action from a plurality of pre-defined authentication actions. Subsequently, the computing platform may generate commands directing an account portal computing platform to allow access, conditionally allow access, or prevent access based on the selected authentication action. Then, the computing platform may send the commands to the account portal computing platform.

Processing authentication requests to secured information systems based on machine-learned user behavior profiles

Aspects of the disclosure relate to processing authentication requests to secured information systems using machine-learned user-account behavior profiles. A computing platform may receive an authentication request corresponding to a request for a user of a client computing device to access one or more secured information resources associated with a user account. The computing platform may capture one or more behavioral parameters and activity data associated with one or more interactions with one or more non-authenticated pages. Then, the computing platform may evaluate the one or more behavioral parameters and the activity data using a behavioral profile associated with the user account. Based on this evaluation, the computing platform may identify the authentication request as malicious and may generate and send one or more denial-of-access commands to prevent the client computing device from accessing the one or more secured information resources associated with the user account.

Processing authentication requests to secured information systems using machine-learned user-account behavior profiles

Aspects of the disclosure relate to preventing unauthorized access to secured information systems. A computing platform may receive, from an end user desktop computing device, a request to login to a user account associated with a user account portal. In response to receiving the request, the computing platform may generate an authentication token in an authentication database and may send a notification to at least one registered device linked to the user account. After sending the notification, the computing platform may receive, from the at least one registered device, an authentication response message. If the authentication response message indicates that valid authentication input was received, the computing platform may update the authentication token to indicate that the request to login to the user account has been approved. After updating the authentication token, the computing platform may provide, to the end user desktop computing device, access to a portal interface.

Preventing unauthorized access to secured information systems using authentication tokens and multi-device authentication prompts

Aspects of the disclosure relate to determining optimal machine learning models. A computing platform may collect, via a network, data indicative of login activity to an enterprise resource. One or more initial features indicative of the login activity may be identified. Based on the one or more initial features, an initial test dataset and a test model may be generated. Then, the test model may be deployed in a production environment. Subsequently, the computing platform may identify one or more production feature vectors. Then, the computing platform may generate, based on the one or more production feature vectors, a training dataset. Subsequently, the computing platform may perform, for the training dataset, a data quality check. Then, the computing platform may predict, by applying Bayesian optimization to the training dataset, an optimal machine learning model. Subsequently, the computing platform may apply the optimal machine learning model to detect unauthorized activity.

Determining optimal machine learning models

Aspects of the disclosure relate to linking channel-specific systems with a user authentication hub. In some embodiments, a computing platform may receive, from a telephone agent support computer system associated with a telephone agent channel, an authentication request for a user account. The computing platform may generate a set of one or more authentication prompts based on a set of authentication rules defined for the telephone agent channel and may provide the set of one or more authentication prompts generated based on the set of authentication rules defined for the telephone agent channel. Subsequently, the computing platform may validate one or more responses to the set of one or more authentication prompts. Based on validating the one or more responses, the computing platform may provide user account information associated with the user account to the telephone agent support computer system associated with the telephone agent channel.

Linking channel-specific systems with a user authentication hub to provide omni-channel user authentication

Aspects of the disclosure relate to linking channel-specific systems with a user authentication hub. In some embodiments, a computing platform may receive, from a mobile device, an authentication request for a user account and may generate one or more authentication prompts. Subsequently, the computing platform may provide the one or more authentication prompts and may validate responses to the authentication prompts. Based on validating the responses to the authentication prompts, the computing platform may generate one or more security questions based on historical information associated with the user account. Subsequently, the computing platform may provide the one or more security questions and may validate one or more responses to the one or more security questions. Based on validating the one or more responses to the one or more security questions, the computing platform may provide user account information associated with the user account to the mobile device.

Processing authentication requests to secured information systems based on user behavior profiles

Aspects of the disclosure relate to processing authentication requests to secured information systems based on machine-learned event profiles. A computing platform may receive an authentication request corresponding to a request for a user of a client computing device to access one or more secured information resources associated with a user account in a client portal session. The computing platform may capture one or more behavioral parameters and may generate one or more authentication prompts. Thereafter, the computing platform may receive one or more authentication prompt responses and may evaluate an event pattern. Based on evaluating the event pattern and validating the one or more authentication prompt responses, the computing platform may generate and send one or more authentication commands directing an account portal computing platform to allow access to the one or more secured information resources associated with the user account in the client portal session.

Processing authentication requests to secured information systems based on machine-learned event profiles

Aspects of the disclosure relate to processing authentication requests to secured information systems using machine-learned user-account behavior profiles. A computing platform may receive an authentication request corresponding to a request for a user of a client computing device to access one or more secured information resources associated with a user account. The computing platform may capture one or more behavioral parameters and may authenticate the user of the client computing device to the user account based on the one or more behavioral parameters and one or more authentication credentials. The computing platform then may generate and send one or more authentication commands directing an account portal computing platform to allow access to the one or more secured information resources. Subsequently, the computing platform may capture activity data associated with one or more interactions in a client portal session and may update a behavioral profile associated with the user account.

Systems for detecting unauthorized user and controlling dynamic user interface functionality are provided. The system may receive a request to access functionality that may include login credentials of a user. The request may also include additional information associated with a computing device from which the request is received. The request and additional data may be analyzing using one or more machine learning datasets to determine whether a user requesting access is an authorized user or an unauthorized user. If the user is an authorized user, the user may be authenticated to the system an authentic user interface having enabled functionality may be generated. If the user is an unauthorized user, a decoy user interface having functionality disabled may be generated.

Controlling dynamic user interface functionality using a machine learning control engine

Growing community of inventors

Issaquah, WA, United States of America

Xianhong Zhang