Patents by Inventor Purva SHANKER
Purva SHANKER 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: 20250055865Abstract: Systems and methods for detecting unauthorized access are disclosed herein. In some aspects, the system receives a combined activity dataset. The system updates a base breach detection model based on the combined activity dataset to generate a combined breach detection model. The system duplicates the combined breach detection model to generate a first breach detection model for a first user. The system receives a first activity dataset for the first user and trains a labeling model to associate activities from the first activity dataset with the first user. The system processes the combined activity dataset using the labeling model to associate activities from a portion of the combined activity dataset with the first user. The system updates the base breach detection model based on activities from the first activity dataset and the portion of the combined activity dataset to generate a second breach detection model to detect breach activity.Type: ApplicationFiled: August 7, 2023Publication date: February 13, 2025Applicant: Capital One Services, LLCInventors: Joshua EDWARDS, Purva SHANKER, Nathan WOLFE
-
Patent number: 12216684Abstract: A method includes obtaining a first dataset associated with a first set of classes and a second dataset associated with a second set of classes, where the first and second datasets share features. The method also includes creating an aggregated dataset having a quality score for the second set of classes that satisfy a first threshold. Creating the aggregated dataset includes determining a first and second set of clusters based on the first and second datasets. The method includes determining a distance based on a record of the first set of clusters and a second record of the second set of clusters in a feature space. The method further includes determining a set of analysis scores based on the distance and generating an indication for a set of records associated with the set of analysis scores based on the set of analysis scores satisfying a second threshold.Type: GrantFiled: August 19, 2022Date of Patent: February 4, 2025Assignee: Capital One Services, LLCInventors: Nathan Wolfe, Purva Shanker, Joshua Edwards, Gang Mei
-
Patent number: 12210599Abstract: Systems and methods for providing variable and temporary access to account content for a user in a secured manner through the use of an access token with variable properties are described. The systems and methods provide improved navigability to account content accessed via the access token through the customization of user interfaces. For example, the system and methods may generate user interface templates that comprise a recommended selection and organization of user input fields and/or user interface pages.Type: GrantFiled: July 13, 2022Date of Patent: January 28, 2025Assignee: CAPITAL ONE SERVICES, LLCInventors: Lin Ni Lisa Cheng, Purva Shanker, Sara Margaret Hughes, Allison Trager, Shaun Kieran Webb
-
Publication number: 20250028637Abstract: Systems and methods for a predictive cache of machine learning models. In some aspects, the system, in response to a user request, trains one or more machine learning models using a training dataset and inserts the machine learning models into a user library. The system, without input from the user, based on a distribution of values for a first feature in the training dataset, generates a search request for labeled data having a distribution of values within a threshold of the distribution of values of the first feature. The system adds to the training dataset values from the labeled data to generate an updated training dataset. The system trains a proximate machine learning model based on the updated training dataset and inserts the proximate machine learning model into a predictive cache of machine learning models.Type: ApplicationFiled: July 17, 2023Publication date: January 23, 2025Applicant: Capital One Services, LLCInventors: Tania CRUZ MORALES, Purva SHANKER, Shannon YOGERST, Haytham YAGHI
-
Patent number: 12182089Abstract: Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.Type: GrantFiled: January 18, 2023Date of Patent: December 31, 2024Assignee: Capital One Services, LLCInventors: Tania Cruz Morales, Purva Shanker, Shannon Yogerst, Ignacio Espino, Dan Lin, Nathan Wolfe
-
Publication number: 20240396857Abstract: Methods and systems for reducing network traffic between a service and client devices. In some aspects, the system receives a first data stream for a first type of communication between a service and a client device for a user. In response to determining that the first data stream includes an unresolved user query, the system determines that a second type of communication occurred between the service and the user. The system processes a second data stream for the second type of communication to determine that the second type of communication includes the unresolved user query and a service response to the unresolved user query. The system provides the unresolved user query and the service response to update a machine learning model used by the service to generate service responses to one or more user queries during a future communication of the first type.Type: ApplicationFiled: June 6, 2024Publication date: November 28, 2024Applicant: Capital One Services, LLCInventors: Shannon YOGERST, Purva SHANKER, Tania CRUZ MORALES, Haytham YAGHI
-
Patent number: 12147411Abstract: Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.Type: GrantFiled: January 18, 2023Date of Patent: November 19, 2024Assignee: Capital One Services, LLCInventors: Tania Cruz Morales, Purva Shanker, Shannon Yogerst, Ignacio Espino, Dan Lin, Nathan Wolfe
-
Patent number: 12088546Abstract: Methods and systems for reducing network traffic between a service and client devices. In some aspects, the system receives a first data stream for a first type of communication between a service and a client device for a user. In response to determining that the first data stream includes an unresolved user query, the system determines that a second type of communication occurred between the service and the user. The system processes a second data stream for the second type of communication to determine that the second type of communication includes the unresolved user query and a service response to the unresolved user query. The system provides the unresolved user query and the service response to update a machine learning model used by the service to generate service responses to one or more user queries during a future communication of the first type.Type: GrantFiled: February 15, 2023Date of Patent: September 10, 2024Assignee: Capital One Services, LLCInventors: Shannon Yogerst, Purva Shanker, Tania Cruz Morales, Haytham Yaghi
-
Publication number: 20240275748Abstract: Methods and systems for reducing network traffic between a service and client devices. In some aspects, the system receives a first data stream for a first type of communication between a service and a client device for a user. In response to determining that the first data stream includes an unresolved user query, the system determines that a second type of communication occurred between the service and the user. The system processes a second data stream for the second type of communication to determine that the second type of communication includes the unresolved user query and a service response to the unresolved user query. The system provides the unresolved user query and the service response to update a machine learning model used by the service to generate service responses to one or more user queries during a future communication of the first type.Type: ApplicationFiled: February 15, 2023Publication date: August 15, 2024Applicant: Capital One Services, LLCInventors: Shannon YOGERST, Purva SHANKER, Tania CRUZ MORALES, Haytham YAGHI
-
Publication number: 20240241872Abstract: Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.Type: ApplicationFiled: January 18, 2023Publication date: July 18, 2024Applicant: Capital One Services, LLCInventors: Tania CRUZ MORALES, Purva SHANKER, Shannon YOGERST, Ignacio ESPINO, Dan LIN, Nathan WOLFE
-
Publication number: 20240242115Abstract: Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.Type: ApplicationFiled: January 18, 2023Publication date: July 18, 2024Applicant: Capital One Services, LLCInventors: Tania CRUZ MORALES, Purva SHANKER, Shannon YOGERST, Ignacio ESPINO, Dan LIN, Nathan WOLFE
-
Publication number: 20240241871Abstract: Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.Type: ApplicationFiled: January 18, 2023Publication date: July 18, 2024Applicant: Capital One Services, LLCInventors: Tania CRUZ MORALES, Purva SHANKER, Shannon YOGERST, Ignacio ESPINO, Dan LIN, Nathan WOLFE
-
Publication number: 20240241875Abstract: Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.Type: ApplicationFiled: January 18, 2023Publication date: July 18, 2024Applicant: Capital One Services, LLCInventors: Tania CRUZ MORALES, Purva SHANKER, Shannon YOGERST, Ignacio ESPINO, Dan LIN, Nathan WOLFE
-
Publication number: 20240232699Abstract: Embodiments relate to enhancing data labeling for machine learning model training using context data. The context data is provided to data labelers to improve accuracy of labels assigned by the data labelers, such as those assigned for ambiguous or unclear target objects. In an example, a system determines and provides context sets for target objects at user devices. The system generates a first training dataset and a second training dataset with labels obtained in connection with a first subset and a second subset of the context sets, respectively. The system trains a first and second instance of a machine learning model on the first training dataset and the second training dataset, respectively, and determines a respective accuracy score for the model instances. If the first instance is more accurate than the second instance, the system generates subsequent context sets based on characteristics of the first subset of context sets.Type: ApplicationFiled: January 5, 2023Publication date: July 11, 2024Applicant: Capital One Services, LLCInventors: Purva SHANKER, Shannon YOGERST, Tania CRUZ MORALES, Haytham YAGHI
-
Publication number: 20240089105Abstract: The invention relates to generating cryptographic tokens based on received records. The system may receive a blockchain operation request comprising a plurality of records, encrypt a plurality of records into an encrypted payload, generate a cryptographic token based on the encrypted payload, and cause the cryptographic token to be committed to a blockchain.Type: ApplicationFiled: September 12, 2022Publication date: March 14, 2024Applicant: Capital One Services, LLCInventors: Julian DUQUE, Purva SHANKER, Samuel RAPOWITZ, Dennis LIU
-
Publication number: 20240086549Abstract: The invention relates to generating cryptographic tokens based on characteristic determination. The system may generate and transmit a request for encrypted records from a blockchain node, receive the encrypted records from the blockchain node, decrypt the encrypted records into payload identifiers, retrieve sets of item identifiers based on the payload identifiers, determine characteristics associated with the sets of items, and generate a new cryptographic token that indicates the characteristics.Type: ApplicationFiled: September 12, 2022Publication date: March 14, 2024Applicant: Capital One Services, LLCInventors: Samuel RAPOWITZ, Julian DUQUE, Purva SHANKER, Dennis LIU
-
Publication number: 20240061867Abstract: A method includes obtaining a first dataset associated with a first set of classes and a second dataset associated with a second set of classes, where the first and second datasets share features. The method also includes creating an aggregated dataset having a quality score for the second set of classes that satisfy a first threshold. Creating the aggregated dataset includes determining a first and second set of clusters based on the first and second datasets. The method includes determining a distance based on a record of the first set of clusters and a second record of the second set of clusters in a feature space. The method further includes determining a set of analysis scores based on the distance and generating an indication for a set of records associated with the set of analysis scores based on the set of analysis scores satisfying a second threshold.Type: ApplicationFiled: August 19, 2022Publication date: February 22, 2024Applicant: Capital One Services, LLCInventors: Nathan WOLFE, Purva SHANKER, Joshua EDWARDS, Gang MEI
-
Publication number: 20240020570Abstract: Systems and methods for providing variable and temporary access to account content for a user in a secured manner through the use of an access token with variable properties are described. The systems and methods provide improved navigability to account content accessed via the access token through the customization of user interfaces. For example, the system and methods may generate user interface templates that comprise a recommended selection and organization of user input fields and/or user interface pages.Type: ApplicationFiled: July 13, 2022Publication date: January 18, 2024Applicant: Capital One Services, LLCInventors: Lin Ni Lisa CHENG, Purva SHANKER, Sara Margaret HUGHES, Allison TRAGER
-
Publication number: 20240020365Abstract: Systems and methods for providing variable and temporary access to account content for a user in a secured manner through the use of an access token with variable properties are described. The systems and methods provide improved navigability to account content accessed via the access token through the customization of user interfaces. For example, the system and methods may generate user interface templates that comprise a recommended selection and organization of user input fields and/or user interface pages.Type: ApplicationFiled: July 13, 2022Publication date: January 18, 2024Applicant: Capital One Services, LLCInventors: Lin Ni Lisa CHENG, Purva SHANKER, Sara Margaret HUGHES, Allison TRAGER, Shaun Kieran WEBB
-
Patent number: 11847117Abstract: A data model identifying a first and second table may be stored, the first table comprising a first and second attribute, the second table comprising a third attribute. A first filter parameter of a first filter and a second filter parameter of a second filter may be obtained. A first tag value may be associated with the first and second filters. A set of filters including the first and second filters may be determined in response to a determination that the first and second filters are associated with the first tag value. An argument indicating the first and second filter parameters may be generated based on the set of filters. A call to the first table may be executed based on the argument, the execution of the call causing values of the first and second attributes to be obtained based on the first and second filter parameters.Type: GrantFiled: July 14, 2021Date of Patent: December 19, 2023Assignee: Capital One Services, LLCInventor: Purva Shanker