Capital One Services, LLC

Hazar Alsahnawi

Shannon Yogerst

Dan Lin

Austin Erickson

John E Jones, Iv

Michael Bakr Mossoba

Leeyat Bracha Tessler

Mary Sweeney

Dennis Liu

Alexander Mireles

Benjamin Eng

A method and related system of operations that include retrieving a set of on-chain transactions of an on-chain address from a scan of on-chain transactions based on a request identifying the on-chain address and generating a sequence of embedding vectors by providing the set of on-chain transactions to a first machine learning model. The method further includes predicting a first use pattern of the on-chain address by providing the sequence of embedding vectors to a second machine learning model and determining whether the first use pattern matches with a second use pattern. The method further includes updating a record of the on-chain address in an off-chain database to indicate anomalous activity in response to a determination that the first use pattern does not match with the second use pattern.

Machine learning to detect misclassified on-chain addresses

Systems and methods for preventing blockchain functions associated with malicious entities are disclosed herein. A first blockchain status update may be received from a first authorizing entity. A first counter and a first plurality of authorizing entities can be determined. The first counter and the first plurality of authorizing entities can be updated. The first counter can be compared to a consensus threshold. A first instruction to block requests for communications to the first entity address can be generated. A confirmation of the first instruction can be generated for display.

Systems and methods for preventing communications with malicious cryptography-based storage application addresses using consensus

Systems and methods for verifying digitally secured communications between users are disclosed herein. A first user identifier for a first user may be received. A first blockchain address for the first user based on the first user identifier may be determined. A first non-transferable token corresponding to the first user may be generated. A request for a communication may be received. A target address and a target user identifier may be extracted. Whether the communication metadata matches the token metadata may be determined. A validation status for the communication may be determined.

Systems and methods for verifying cryptographically secured communications between users using non-transferable tokens

In some implementations, a device may obtain video and/or lidar data relating to a physical card captured while the physical card is flipped from a first side of the physical card to a second side of the physical card. The device may determine an authenticity of the physical card, where the authenticity of the physical card is based at least in part on one or more physical characteristics of the physical card identifiable from the video and/or the lidar data. The one or more physical characteristics may relate to a thickness of the physical card, a length and width of the physical card, a shape of the physical card, and/or a design feature of the physical card. The device may cause, based on a determination that the physical card is authentic, information associated with the physical card to be transmitted or stored to be used in connection with a transaction.

Authenticating a physical card using sensor data

Growing community of inventors

McLean, VA, United States of America

Hazar Alsahnawi