Capital One Services, LLC

Shabnam Kousha

Joshua S Edwards

Tyler Maiman

Asher Smith-Rose

Lin Ni Lisa Cheng

Daniel E Miller

Samuel Rapowitz

Joshua Peters

Haytham Yaghi

Evelyn Burke

Caroline Williams

Maxwell Gotsch

In some embodiments, the present disclosure provides an exemplary method that may include steps of obtaining, a permission from each user to monitor a plurality of activities; continually receiving monitoring data of the plurality of activities; identifying a plurality of related incoming interaction sessions within the predetermined period of time; automatically verifying at least one common session parameter; utilizing a machine learning algorithm to determine at least one transaction; determining a plurality of fraudulent transactional rules; automatically adjusting the plurality of fraudulent transactional rules; utilizing a natural language processing algorithm to automatically generate at least one interaction notification; receiving a response to the interactive notification; and automatically updating the database of known session interaction parameters based on the response to the interactive notification.

Computer-based systems configured to generate an interactive communication for a plurality of the repetitive interaction sessions and methods of use thereof

In some embodiments, the present disclosure provides an exemplary method that may include steps of obtaining a permission from the user to monitor a plurality of activities executed within the computing device; continually monitoring the plurality of activities executed within the computing device for a predetermined period of time; identifying an indication of an incoming interaction session within the predetermined period of time; verifying at least one session interaction parameter associated with the incoming interaction session to identify the incoming interaction session as a repeat interaction session; dynamically retrieving at least one relation-specific notation from a plurality of relation-specific notations to display; instructing an input GUI element to display input data associated with the repeat interaction; automatically updating the plurality of relation-specific notations associated with the historical data relationship; and instructing at least one programmable output GUI to display a notification identifying the update of the plurality of relation-specific notations.

Computer-based platforms and systems configured to enhance a call screen with user-controlled notes associated with the call and methods of use thereof

Systems and methods of caller identification differentiation via machine learning techniques are disclosed. In one embodiment, an exemplary computer-implemented method may include: receiving a permission indicator identifying a permission by the user to detect calls being received by a computing device; receiving an indication of a current call being received; utilizing a trained call differentiation machine learning model to determine a likelihood that the current call is of a first call type or a second call type, where the first call type is associated with a first type of activity and the second call type is associated with a second type of activity.

Computer-based systems configured for caller identification differentiation and methods of use thereof

Systems and methods of the present disclosure enable intelligent dynamic caching of data by accessing an activity history of historical electronic activity data entries associated with a user account, and utilizing a trained entity relevancy machine learning model to predict a degree of relevance of each entity associated with the historical electronic activity data entries in the activity history based at least in part on model parameters and activity attributes of each electronic activity data entry. A set of relevant entities are determined based at least in part on the degree of relevance of each entity. Pre-cached entities are identified based on pre-cached entity data records cached on the user device, and un-cached relevant entities from the set of relevant entities are identified based on the pre-cached entities. The cache on the user device is updated to cache the un-cached entity data records associated with the un-cached relevant entities.

Automated predictive caching of cloud-sourced data and methods of use thereof

In some embodiments, the present disclosure provides an exemplary method that may include steps of obtaining a trained spam upsurge detection machine learning model that determines when a current frequency associated with spam communications received by a current user exceeds a baseline frequency associated with the current user; receiving a permission indicator identifying a permission by the user to detect communications being received by the computing device; receiving an indication of at least one communication being received; determining the at least one communication as a particular spam communication; updating a frequency at which spam communications have been received by the user based at least in part on the particular spam communication; utilizing the trained spam upsurge detection machine learning model to determine that the frequency exceeds a baseline frequency associated with the user; and initiating a scan of one or more dark web resources.

Computer based systems configured to adjust frequency scanning of the dark web in search of information and method of use thereof

A computer-implemented method comprising: instructing a computing device to obtain a permission from the user to monitor a plurality of activities executed within the computing device; receiving a call session request from a computing device associated with a second user; utilizing a trained machine learning algorithm to confirm an identity of the user based on the call session interaction parameter to confirm the call interaction session as a verified call interaction session; generating a fictitious call session interaction parameter that differs from the at least one call session interaction parameter; modifying the call session request into a verified call session request by replacing the call session interaction parameter with the fictitious call session interaction parameter; utilizing a telecommunication network to transmit the verified call session request with the at least one fictitious session interaction parameter to the first calling-enabled computing device; automatically instructing the first calling-enabled computing device to modify a call session notification GUI to display notification data.

Computer-based systems configured for automatically generating fictitious call session interaction parameter(s) and methods of use thereof

Systems and methods of location-aware caller identification via machine learning techniques are disclosed. In one embodiment, an exemplary computer-implemented method may include: utilizing a trained call annotation machine learning model to determine one or both of an annotating condition and annotating location granularity, and associate one location of a user with one phone number of the user based at least on one or both of the annotating condition and the annotating location granularity; receiving second transactional information of one transaction associated with a first user; extracting second location information from the second transactional information of the one transaction; utilizing the trained call annotation machine learning model to automatically annotate one phone number record of one phone number of the first user, with the second location information at the annotating location granularity to form at least one user-specific location-specific annotated phone number record.

Computing devices configured for location-aware caller identification and methods/systems of use thereof

Disclosed herein are system, method, and computer program product embodiments for training and deploying a machine learning model to generate an assessment of risks and mitigations in response to a novel initiative request. After generating labeled data from a corpus of prior risk assessments, a machine learning model may be trained to programmatically generate a risk assessment in response to a novel initiative request. A risk auditor may subsequently review the risk assessment generated using the machine learning model to provide various feedback reflecting the accuracy of the risks and mitigations. The machine learning model may then be retrained based on the feedback provided by the risk auditor to provide more accurate risks and mitigations in response to future initiatives.

Method for scoring confidence of an algorithmically proposed risk

The present disclosure provides an exemplary method, system, and computing device that may include the steps of receiving a first indication that information of a user has been detected at one or more dark web resources; classifying the item of the compromised information into an information type category; receiving a permission indicator to detect communications by the computing device; receiving a second indication of a communication; receiving a third indication that the user engages an interaction with the communication; instructing the computing device to execute a technique to obtain data for the communication; receiving the data for the communication; determining the communication is a spam communication; determining a current information type category being discussed during the spam communication; making a determination that the current information type category corresponds to the information type category; and instructing a graphical user interface to display an alert to the user.

Computer-based systems configured for contextual notification of monitored dark web intelligence and methods of use thereof

Systems, methods, and non-transitory computer readable medium disclosed herein relate to identity verification and authorization method. In one embodiment, the system can generate and send a message to a device associated with a user based on an initiated request from the user and a determination the user should be authenticated, wherein the message requests a content-based response from the user to authenticate the user. In another embodiment, the system can receive the content-based response from the user in reply to the message, wherein the content-based response comprises SMS (short message service) metadata, emoji, photo, video, audio, or a combination thereof. In another embodiment, the system can authenticate the user based on a determination of a confirmed match between the content-based response from the user and a response key preselected by the user.

Alternative identity verification and authorization method

In some embodiments, the present disclosure provides an exemplary method that may include steps of obtaining a permission to monitor a plurality of activities executed within the second computing device; continuously monitoring the plurality of activities executed within the second computing device for a predetermined period of time; receiving an indication of an incoming interaction session within the predetermined period of time; automatically verifying at least one session interaction parameter associated with the incoming interaction session to identify the incoming interaction session as a suspect interaction session; determining when a duration of time associated with the suspect interaction session meets or exceeds a predetermined duration threshold; utilizing a trained detection machine learning model to model a confidence value associated with the suspect interaction session; identifying at least one future activity subsequent to the suspect interaction session; and automatically instructing the first computing device to halt the future activity.

Computer-based systems configured for utilization of a trained detection machine learning model for activity determination and methods of use thereof

Systems and methods of video password based user authentication via machine learning techniques are disclosed. In one embodiment, an exemplary computer-implemented method may comprise: receiving a request to register a video password from a first user; establishing at least one user-specific authentication criterion for the first user based on first video password data; receiving a login attempt from a second computing device associated with a second user who submits to be the first user, the login attempt comprising second video password data; utilizing a trained video password authentication machine learning model to determine a first comparison result and a second comparison result based on the first and second video password data, and accepting or rejecting the login attempt based on at least one of the first comparison result and the second comparison result.

Computer-based systems having components configured/programmed for video password based authentication and methods of use thereof

Disclosed are techniques and apparatuses that are configured to receive an indication that a web browsing session executing on an enterprise server needs additional information based on a request for additional information being sent to a client device. The request may include an identifier of the web browsing session and an identifier of an enterprise server that initiated the web browsing session. A globally unique identifier related to the web browsing session and an identifier of the enterprise server is stored in a common data store. The web browsing session may be paused when the web browsing session requests additional information from a client device. The client device may respond with the additional information. The system may provide the identifier of the enterprise server to a load balancing component so the identified web browsing session executing on the enterprise server may continue to be used.

Scalable server-based web scripting with user input

Systems and methods of anti-vishing OTP protection via machine learning techniques are disclosed. In one embodiment, an exemplary computer-implemented method may comprise: receiving a permission indicator identifying a permission by the user to detect OTPs and calls being received by a computing device; receiving an indication of an OTP data item being received; processing the OTP data item to determine a time duration during which a particular OTP included therein is valid; utilizing a trained OTP protection machine learning model to determine phone number(s) as presenting a security risk with respect to the OTP data item; and instructing the computing device to commence at least one security measure based at least in part on a contact list updated with an indication that the phone number(s) present a security risk with regard to the particular OTP during the time duration of the particular OTP.

Growing community of inventors

Washington, DC, United States of America

Shabnam Kousha