Patents by Inventor Abhishek TEWARI

Abhishek TEWARI 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).

  • Publication number: 20250165442
    Abstract: The systems and methods for improved precision in detecting false-positive changepoints in time-series data. The system may receive a first dataset of time-series datapoints. The system may determine a first changepoint in the first dataset. The system may determine a first category of known values for the first dataset. The system may, based on the first category of the known values, select a first model from a plurality of models for determining whether the first changepoint corresponds to a first false-positive changepoint. The system may, in response to selecting the first model, process, using the first model, the first changepoint and a first value of the known values to determine a first output. The system may generate for display, in a user interface, a first recommendation based on the first output, wherein the first recommendation indicates whether the first changepoint corresponds to the first false-positive changepoint.
    Type: Application
    Filed: January 2, 2024
    Publication date: May 22, 2025
    Applicant: Capital One Services, LLC
    Inventors: Zhengqing LIU, Anupam SHIT, Abhishek TEWARI, Hassan SHALLAL
  • Publication number: 20240386287
    Abstract: Systems, methods, articles of manufacture, and computer program products to generate decision trees are described. In some embodiments, a computer-implemented method to generate a machine learning decision tree model may include, via at least one processor of a computing device, determining a set of numeric variables for each of the plurality of categorical variables, determining an event rate for each of the plurality of categorical variables, determining, using training data, a plurality of splits for assigning the plurality of categorical variables to nodes of the machine learning decision tree model, wherein the plurality of splits comprises a plurality of multi-categorical splits assigning multiple of the plurality of categorical variables to a single node based on the event rate, and accessing data to generate the machine learning decision tree model comprising a plurality of nodes, at least a portion of the nodes assigned one of the plurality of multi-categorical splits.
    Type: Application
    Filed: November 3, 2023
    Publication date: November 21, 2024
    Applicant: Capital One Services, LLC
    Inventors: Prashanta SAHA, Joydeep DASGUPTA, Abhishek TEWARI, Arun Kaushik Narmadha RAMESH, Arindam Roy CHOWDHURY, Anupam SHIT, Gaurav KEDIA
  • Publication number: 20240386333
    Abstract: Systems and methods for generating decision trees are described. In some embodiments, a computer-implemented method to generate a machine learning decision tree may include accessing data to generate the decision tree, the data comprising a plurality of categorical variables; determining an event rate for each of the plurality of categorical variables generating a recursive tree comprising one or more nodes for the plurality of categorical variables via, for each of the one or more nodes: determining a node split for the one or more nodes based on the event rate for each of the plurality of category variables, the node split to split each of the one or more nodes into two nodes, and determine an encoder based on the node split for each level of the recursive tree; and generating the decision tree using a plurality of encoders comprising the encoder for each level of the recursive tree.
    Type: Application
    Filed: November 3, 2023
    Publication date: November 21, 2024
    Applicant: Capital One Services, LLC
    Inventors: Prashanta SAHA, Anupam SHIT, Abhishek TEWARI, Gaurav KEDIA, Joydeep DASGUPTA, Arun Kaushik Narmadha RAMESH
  • Patent number: 12020319
    Abstract: Systems, methods, and apparatuses for automatically detecting the risk of dealer default are described. A machine learning model may be trained using a gradient boosting technique on a dataset that includes historical lien information, historical dealer information, and historical vehicle default information. Lien information and financing information for vehicles may be received and correlated. Furthermore, a machine learning model may generate aggregate risk scores for dealers. Furthermore, determination of an aggregate risk score that exceeds a threshold risk score may cause the generation of a notification.
    Type: Grant
    Filed: June 27, 2022
    Date of Patent: June 25, 2024
    Assignee: Capital One Services, LLC
    Inventors: Pedro Partiti De Oliveira, Marrian Kitchens, Papa D. Ndiaye, Joseph T. Allison, Collin Longmire, John Runge, Dong Ji, Abhishek Tewari, Bing Liu
  • Publication number: 20230260018
    Abstract: Systems, methods, and apparatuses for automatically detecting the risk of dealer default are described. A machine learning model may be trained using a gradient boosting technique on a dataset that includes historical lien information, historical dealer information, and historical vehicle default information. Lien information and financing information for vehicles may be received and correlated. Furthermore, a machine learning model may generate aggregate risk scores for dealers. Furthermore, determination of an aggregate risk score that exceeds a threshold risk score may cause the generation of a notification.
    Type: Application
    Filed: June 27, 2022
    Publication date: August 17, 2023
    Inventors: Pedro Partiti De Oliveira, Marrian Kitchens, Papa D. Ndiaye, Joseph T. Allison, Collin Longmire, John Runge, Dong Ji, Abhishek Tewari, Bing Liu
  • Publication number: 20230260019
    Abstract: Systems, methods, and apparatuses for automatically prioritizing unperfected liens are described. A machine learning model may be trained to detect a risk of dealer default. Lien information corresponding to liens for vehicles associated with dealers may be received. Prioritized lien information may be generated based on selection of the liens that satisfy criteria. Financing information for vehicles may be received and correlated with the prioritized lien information. Furthermore, the machine learning model may be used to generate risk scores associated with the liens. The liens may be prioritized based on the aggregate risk scores.
    Type: Application
    Filed: June 27, 2022
    Publication date: August 17, 2023
    Inventors: Pedro Partiti De Oliveira, Collin Longmire, Marrian Kitchens, Joseph T. Allison, Papa D. Ndiaye, John Runge, Dong Ji, Abhishek Tewari, Bing Liu
  • Publication number: 20220301022
    Abstract: Methods and systems are described for improvements to the use of distributed computer networks. For example, conventional systems may rely on the distribution of network or application traffic across multiple servers and may maintain load balancers to maintain that distribution in an efficient manner. Each load balancer may sit between client devices and backend servers, receiving and then distributing incoming requests to any available server capable of fulfilling them. The load balancers may ensure that no one server is overworked based on the number of processing requests directed to that server, which could degrade performance.
    Type: Application
    Filed: March 19, 2021
    Publication date: September 22, 2022
    Applicant: Capital One Services, LLC
    Inventors: Cheng JIANG, Yun ZHOU, Christy PURNADI, Abhishek TEWARI, Seyed Hossein Zahed ZAHEDANI, Ryan SRALLA, Yasong ZHOU, Brian McGILL, Gavin OLSON
  • Publication number: 20220300364
    Abstract: Methods and systems are for generating real-time resolutions of errors arising from user submissions, computer processing tasks, etc. For example, the methods and systems described herein recite improvements for detecting errors in one or more user submissions and providing resolutions in real-time. To provide these improvements, the methods and systems use a machine learning model that is trained to return probability error scores based on a plurality of variables. By using the multivariate approach, the methods and systems may produce a highly accurate detection.
    Type: Application
    Filed: March 17, 2021
    Publication date: September 22, 2022
    Applicant: Capital One Services, LLC
    Inventors: Vinuta NAGARADDI, Shelly BERGMAN, Abhishek TEWARI, Prableen KAUR, Yasong ZHOU, Xiaolong BAO, Andrew COZZOLINO, Soma Sekhar TANKALA, Junli YUAN, Gowtham GUDURU
  • Publication number: 20220301021
    Abstract: Methods and systems are described for improvements to the use of distributed computer networks. For example, conventional systems may rely on the distribution of network or application traffic across multiple servers and may maintain load balancers to maintain that distribution in an efficient manner. Each load balancer may sit between client devices and backend servers, receiving and then distributing incoming requests to any available server capable of fulfilling them. The load balancers may ensure that no one server is overworked based on the number of processing requests directed to that server, which could degrade performance.
    Type: Application
    Filed: March 19, 2021
    Publication date: September 22, 2022
    Applicant: Capital One Services, LLC
    Inventors: Cheng Jiang, Yun Zhou, Christy Purnadi, Abhishek Tewari, Seyed Hossein Zahed Zahedani, Ryan Sralla, Yasong Zhou, Brian McGill, Gavin Olson
  • Publication number: 20220301020
    Abstract: Methods and systems are described for improvements to the use of distributed computer networks. For example, conventional systems may rely on the distribution of network or application traffic across multiple servers and may maintain load balancers to maintain that distribution in an efficient manner. Each load balancer may sit between client devices and backend servers, receiving and then distributing incoming requests to any available server capable of fulfilling them. The load balancers may ensure that no one server is overworked based on the number of processing requests directed to that server, which could degrade performance.
    Type: Application
    Filed: March 19, 2021
    Publication date: September 22, 2022
    Applicant: Capital One Services, LLC
    Inventors: Cheng JIANG, Yun ZHOU, Christy PURNADI, Abhishek TEWARI, Seyed Hossein Zahed ZAHEDANI, Ryan SRALLA, Yasong ZHOU, Brian McGILL, Gavin OLSON
  • Publication number: 20220277387
    Abstract: Disclosed herein are system, method, and computer program product embodiments for generating optimized data structure objects. In a given embodiment, a server receives a request to identify a set of data structure objects from a plurality of data structure objects for a specified vehicle and a specified user and based on user objective data and an attribute. The server may generate the plurality of data structure objects within constraints of the user objective data, policy data, and the attribute. The server may identify the set of data structure objects from the plurality of data structure objects. Each data structure object of the set is within a threshold distance to a value of at least one dimension of a set of the dimensions of the set of dimensions of a model.
    Type: Application
    Filed: February 26, 2021
    Publication date: September 1, 2022
    Applicant: Capital One Services, LLC
    Inventors: Drake R. SANDERSON, Rukmani THIRUPPATHI, Adam GERNES, Krishna Mohan BIBIREDDY, Taylor Elliott GERON, Thomas BUSATH, Jonathan LAI, Abhishek TEWARI