Capital One Services, LLC

Other

Sierra Nevada Corporation

Joshua S Edwards

Michael Bakr Mossoba

Abdelkader M'Hamed Benkreira

Tyler Maiman

Adam Vukich

Samuel Rapowitz

Viraj Chaudhary

David Septimus

David Gabriele

Jenny Melendez

Vu Nguyen

Ljubica Chatman

Lin Ni Lisa Cheng

Jonatan Yucra Rodriguez

Abdelkader Benkreira

Shabnam Kousha

Jason Ji

Mykhaylo Bulgakov

Eric Loucks

Alexandra Colevas

George Bergeron

Lukiih Cuan

Asher Smith-Rose

Kevin Grant Osborn

Andrea Montealegre

Salik Shah

Alexander Lin

Daniel Miller

Colin Hart

Max Miracolo

Abdelkadar M'Hamed Benkreira

Bryant Alan Yee

Daniel John Marsch

Vyjayanthi Vadrevu

Kaitlin J Newman

Daniel E Miller

Phoebe Atkins

Carlos Eduardo Rodriguez

Molly Elizabeth Johnson

Angelina Wu

Mia Rodriguez

Cruz Vargas

Jiawei Zhao

Rajko Ilincic

Latika Gulati

Aysu Ezen Can

Jennifer Kwok

Tania Cruz Morales

Francisco Perezleon

Carrie Norman

Feng Qiu

Guadalupe Bonilla

Meredith Leigh Critzer

Sarah Jane Cunningham

Kyle Rankin Johnson

William Carroll

Gaurang J Bhatt

Abhay Donthi

Kathryn Tikoian

Abdelkader M'hamed Benkreira

Brian McClanahan

Brian Lee

Vahid Khanagha

Jason Zwierzynski

Sara Rose Brodsky

Mohammad Shami

Mark Rothbaum

Bogdan Nedanov

Conor Anstett

Daniel Herrington

Sophie Bermudez

Austen Novis

Jacob Learned

Michael Saia

Jerry Wagner

Abdelkader M′Hamed Benkreira

James Zarakas

Kevan Emmott

Purva Shanker

Matthew Carson

Nathan Wolfe

Dan Lin

Eric Medin

Abdelkadar M′Hamed Benkreira

Noel Lyles

Lauren Fox

Lea Cody

Sarvani Kare

Erik Neighbour

Louis Buell

Shaun Webb

Peter Fischer

Haytham Yaghi

Jesse Emery

Marcelo Jabif Epsztejn

Samuel Yip

Justin Anthony Bowers

Alexander Mireles

Mary Sweeney

In some implementations, a user device may receive a set of words, for conversion to speech, including a plurality of subsets of the set of words, where each subset of the plurality of subsets is associated with one or more corresponding formatting properties. The user device may identify, within the set of words, a first subset of the plurality of subsets as relevant based on the one or more corresponding formatting properties associated with the first subset. Additionally, the user device may identify, within the set of words, a second subset of the plurality of subsets as not relevant based on the one or more corresponding formatting properties associated with the second subset. Accordingly, the user device may input the first subset to a text-to-speech algorithm.

Flexible text-to-speech for screen readers

In some implementations, a user device may receive, from a first other device, a first message, wherein the first message includes information indicating an event associated with an account of a user of the user device. The user device may provide, based on receiving the first message, a virtual keyboard for display on a display screen of the user device. The user device may obtain, based on providing the virtual keyboard, particular input. The user device may send to a second other device, and based on the particular input, a second message, wherein sending the second message is to cause the second other device to perform one or more actions associated with the event that is associated with the account of the user.

Using a virtual keyboard to enter particular input

Systems and methods for identifying data labels for submitting to additional data labeling routines based on embedding clusters are described. For example, the system may retrieve an unlabeled dataset, wherein the unlabeled dataset comprises a plurality of unlabeled samples. The system may generate a plurality of embeddings based on the unlabeled dataset, wherein the plurality of embeddings comprises a respective embedding for each unlabeled sample in the plurality of unlabeled samples. The system may cluster the plurality of embeddings into a plurality of clusters. The system may, based on the first cluster, determine a first labeled sample of the first labeled dataset corresponding to the first cluster and determine to submit the first labeled sample to a second data labeling routine. The system may generate, using the second data labeling routine, a second labeled sample based on the first labeled sample.

Systems and methods for identifying data labels for submitting to additional data labeling routines based on embedding clusters

Systems and methods for improvements to multifactor authentication (MFA) techniques are described herein. The improved and enhanced MFA techniques can leverage user interaction data to predict an entity-issued user-specific identification document that an authorized user of an authorized user account is most likely to carry on their person. Upon a request to access an authorized user account being received, the improved and enhanced MFA techniques can request that the user capture, in real-time, an image depicting the predicted identification document. Using additional artificial intelligence, a determination may be made as to whether the image depicts a real instance of the entity-issued user-specific identification document. This determination may result in access to the authorized user account being provided or denied.

Identification card based multi-factor authentication

Systems and methods of the present disclosure enable automated identification of restrictions on reversals of data entries by receiving location data from a computing device associated with a user, and utilizing a data profile classification machine learning model to classify a particular data profile according to a data profile classification type based at least in part on a history of data entries associated with the particular data profile when the physical location is within a predetermined proximity of another physical location associated with the particular data profile. A reversal rate of data entries in the history of data entries is determined for the particular data profile. An electronic activity reversal restriction is determined where the reversal rate is below a predetermined value, and a pop-up notification is presented on the computing device notifying the user of the electronic activity reversal restriction of the particular data profile.

Computer-based systems configured for identifying restricted reversals of operations in data entry and methods of use thereof

A card for storing both emergency information and payment information is disclosed. The card can be used by an emergency responder to retrieve emergency information for the user. The emergency information can enable the emergency responder to better render aid to the user. The card can have one or more processors, a sensor, and a memory in communication with the one or more processors and storing an application thereon. The application can cause the card to receive a request from a user device (e.g., from an emergency medical technician, or EMT) to access emergency information stored in the memory, transmit a security challenge to the user device, authenticate a received response to the security challenge from the user device, and transmit the emergency information to the user device.

Devices and methods for providing emergency information using a payment card

Transaction card

A system may include processor(s), and memory in communication with the processor(s) and storing instructions configured to cause the system to correct ASR errors. The system may receive a transcription comprising transcribed word(s) and may determine whether the transcribed word(s) exceed associated predefined confidence level(s). Responsive to determining a transcribed word does not exceed a predefined confidence level, the system may generate a predicted word. The system may calculate a distance between numerical representations of the transcribed word and the predicted word and may determine whether the distance exceeds a predefined threshold. Responsive to determining the distance exceeds the predefined threshold, the system may determine whether at least one red flag word of a list of red flag words corresponds to a context of the transcription, and, responsive to making that determination, may classify the transcription as associated with a first category.

Systems and methods for correcting automatic speech recognition errors

A fraud detection system may identify a suspected fraudulent transaction based on one or more criteria. The transaction may be associated with a user. Based on identifying a suspected fraudulent transaction, the fraud detection system may receive augmented reality data from an augmented reality device of the user. The fraud detection system may determine, based on the augmented reality data, whether the transaction is fraudulent, and may take an action based on the determination.

Using augmented reality data as part of a fraud detection process

A device may receive customer data, and store the customer data in a secure memory device of the device. The device may implement a backup mode in which the device utilizes the customer data to authorize requests based on determining that the device lacks a network connection. The device may receive, when the device is in the backup mode, a request for currency that includes authentication information. The device, while in the backup mode, may authorize the request for currency based on comparing the authentication information and the customer data. The device may dispense currency based on a result of comparing the authentication information and the customer data. The device may store transaction data associated with dispensing of the currency to permit the transaction data to be transmitted when the network connection is restored.

Automated teller machines (ATMs) having offline functionality

In some implementations, a device may obtain event data relating to events involving one or more entities and one or more individuals, where the one or more entities are associated with entity categories. The device may determine, based on origin locations of the one or more individuals and locations of the one or more entities, distances between the origin locations of the one or more individuals and the locations of the one or more entities, the distances indicating travel distances of the one or more individuals to perform exchanges in the entity categories associated with the one or more entities. The device may determine, based on the event data and the distances, that a geographic area is associated with an opportunity for an entity category. The device may transmit a notification indicating the opportunity for the entity category in the geographic area.

Machine learning model prediction of an opportunity for an entity category in a geographic area

Systems, methods, and apparatuses are described herein for performing sentiment analysis on electronic communications relating to one or more image-based communications methods, such as emoji. Message data may be received. The message data may correspond to a message that is intended to be sent but has not yet been sent to an application. Using a first machine learning model, one or more subsets of the plurality of emoji may be determined. The one or more subsets of the plurality of emoji may comprise one or more different types and quantities of emoji, and may each correspond to the same or a different sentiment. Using a second machine learning model, one or more emojis may be selected from the one or more subsets. The one or more emojis selected may correspond to responses to the message.

Response prediction for electronic communications

Disclosed herein are systems and methods for determining whether an automated telephone call has been answered by a particular user, a different person, or a voicemail service. In response to determining that the call has been answered by the user, a different person, or a voicemail service, the systems and methods can output different respective messages, initiate messaging to the user via alternate communication paths, or perform other actions.

Systems for identifying the answering party of an automated voice call

Systems and methods for external account authentication are disclosed herein. They include receiving a call to pair the external account with a secure account, extracting external data from the external account, the external data corresponding to external account content, providing user activity data from the secure account as an input to an authentication machine learning model, providing the external data as an input to the authentication machine learning model, the authentication machine learning model configured to output a certainty level that the external account is associated with a user of the secure account based on the external data and the activity data, receiving the certainty level from the authentication machine learning model, determining that the certainty level meets a certainty threshold, and pairing the external account with the secure account based on determining that the certainty level meets the certainty threshold.

Systems and methods for external account authentication

Methods and systems are described herein for moving access tokens for validating network operations between user accounts to prevent malicious or nonaligned usage. The system may determine that a user is associated with a first user account and a second user account, retrieve parameters associated with the first and second user accounts, determine that an access token is available for migration, update a binding associated with the access token from the first to the second user account, and, in response to receiving a network operation request, may process the request using the second user account.

Systems and methods for validating network operations between user accounts through access tokens

Growing community of inventors

McLean, VA, United States of America

Joshua S Edwards