Patents by Inventor Selvakumar Albert

Selvakumar Albert has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Data reconciliation system and method
    Publication number: 20260161662
    Abstract: A system assigns a unique prefix to a data table. The system generates a set of document identifications (IDs) for a set of data records included in the data table. The system generates a data key that corresponds to a combination of two or more columns of the data table. The system identifies one or more data records that share a key value with each other, where the key value is associated with the data key. The system determines that the one or more data records belong to a user based at least on identifying that the one or more data records share the key value with each other. The system assigns an entity ID to the identified one or more data records in response to determining that the one or more data records belong to the user.
    Type: Application
    Filed: April 14, 2025
    Publication date: June 11, 2026
    Inventors: Richard William Marker, Amit Gupta, Kripsa Mahesh Mehta, Selvakumar Albert
  • Data reconciliation system and method
    Patent number: 12299000
    Abstract: A system assigns a unique prefix to a data table. The system generates a set of document identifications (IDs) for a set of data records included in the data table. The system generates a data key that corresponds to a combination of two or more columns of the data table. The system identifies one or more data records that share a key value with each other, where the key value is associated with the data key. The system determines that the one or more data records belong to a user based at least on identifying that the one or more data records share the key value with each other. The system assigns an entity ID to the identified one or more data records in response to determining that the one or more data records belong to the user.
    Type: Grant
    Filed: March 16, 2023
    Date of Patent: May 13, 2025
    Assignee: Bank of America Corporation
    Inventors: Richard William Marker, Amit Gupta, Kripsa Mahesh Mehta, Selvakumar Albert
  • Data reconciliation system and method
    Publication number: 20240311393
    Abstract: A system assigns a unique prefix to a data table. The system generates a set of document identifications (IDs) for a set of data records included in the data table. The system generates a data key that corresponds to a combination of two or more columns of the data table. The system identifies one or more data records that share a key value with each other, where the key value is associated with the data key. The system determines that the one or more data records belong to a user based at least on identifying that the one or more data records share the key value with each other. The system assigns an entity ID to the identified one or more data records in response to determining that the one or more data records belong to the user.
    Type: Application
    Filed: March 16, 2023
    Publication date: September 19, 2024
    Inventors: Richard William Marker, Amit Gupta, Kripsa Mahesh Mehta, Selvakumar Albert
  • Neural network host platform for generating automated suspicious activity reports using machine learning
    Patent number: 11593812
    Abstract: Aspects of the disclosure relate to using machine learning techniques for generating automated suspicious activity reports (SAR). A computing platform may generate a labelled transaction history dataset by combining historical transaction data with historical report information. The computing platform may train a convolutional neural network using the labelled transaction history dataset. The computing platform may receive new transaction data and compress the new transaction data using lossy compression. The computing platform may input the compressed transaction data into the convolutional neural network, which may cause the convolutional neural network to output a suspicious event probability score based on the compressed transaction data.
    Type: Grant
    Filed: October 4, 2021
    Date of Patent: February 28, 2023
    Assignee: Bank of America Corporation
    Inventors: Kyle Kreth, Andrew Vlasic, Selvakumar Albert, Jeremiah Thompson, Nancy T. Carrier
  • Neural Network Host Platform for Generating Automated Suspicious Activity Reports Using Machine Learning
    Publication number: 20220027920
    Abstract: Aspects of the disclosure relate to using machine learning techniques for generating automated suspicious activity reports (SAR). A computing platform may generate a labelled transaction history dataset by combining historical transaction data with historical report information. The computing platform may train a convolutional neural network using the labelled transaction history dataset. The computing platform may receive new transaction data and compress the new transaction data using lossy compression. The computing platform may input the compressed transaction data into the convolutional neural network, which may cause the convolutional neural network to output a suspicious event probability score based on the compressed transaction data.
    Type: Application
    Filed: October 4, 2021
    Publication date: January 27, 2022
    Inventors: Kyle Kreth, Andrew Vlasic, Selvakumar Albert, Jeremiah Thompson, Nancy T. Carrier
  • Neural network host platform for generating automated suspicious activity reports using machine learning
    Patent number: 11176557
    Abstract: Aspects of the disclosure relate to using machine learning techniques for generating automated suspicious activity reports (SAR). A computing platform may generate a labelled transaction history dataset by combining historical transaction data with historical report information. The computing platform may train a convolutional neural network using the labelled transaction history dataset. The computing platform may receive new transaction data and compress the new transaction data using lossy compression. The computing platform may input the compressed transaction data into the convolutional neural network, which may cause the convolutional neural network to output a suspicious event probability score based on the compressed transaction data.
    Type: Grant
    Filed: June 2, 2020
    Date of Patent: November 16, 2021
    Assignee: Bank of America Corporation
    Inventors: Kyle Kreth, Andrew Vlasic, Jeremiah Thompson, Nancy T. Carrier, Selvakumar Albert
  • Neural Network Host Platform for Generating Automated Suspicious Activity Reports Using Machine Learning
    Publication number: 20210295338
    Abstract: Aspects of the disclosure relate to using machine learning techniques for generating automated suspicious activity reports (SAR). A computing platform may generate a labelled transaction history dataset by combining historical transaction data with historical report information. The computing platform may train a convolutional neural network using the labelled transaction history dataset. The computing platform may receive new transaction data and compress the new transaction data using lossy compression. The computing platform may input the compressed transaction data into the convolutional neural network, which may cause the convolutional neural network to output a suspicious event probability score based on the compressed transaction data.
    Type: Application
    Filed: June 2, 2020
    Publication date: September 23, 2021
    Inventors: Kyle Kreth, Andrew Vlasic, Jeremiah Thompson, Nancy T. Carrier, Selvakumar Albert