CONTACTLESS TRANSACTION PROTECTION USING GENERATIVE ADVERSARIAL NETWORK ("GAN") DEEP LEARNING TECHNOLOGY

Abstract: methods, apparatus, and systems for identifying and suspending attempted illegitimate transactions. the methods may include attempting a transaction between a contactless transmitter included in a transaction device and a contactless receiver not included in the transaction device. in response to attempting the transaction, the methods may include generating a real-time synthetic data model at a generator module. the methods may include transmitting the real-time synthetic data model to a discriminator module. the methods may include comparing the real-time synthetic data model to identifiers of authorized contactless receivers stored in a knowledge graph. the methods may include executing a prediction whether the attempted transaction will be a legitimate transaction based on the comparison. the methods may include executing the transaction in response to a prediction that the transaction will be a legitimate transaction and terminating the transaction, in response to a prediction that the transaction will be an illegitimate transaction.

Inventor(s): Gowri Sundar Suriyanarayanan, Maneesh Sethia, Abhijit Behera, Shailendra Singh, Aiswary R, Vignesh A S

CPC Classification: G06Q20/4016 ({involving fraud or risk level assessment in transaction processing})

Search for rejections for patent application number 20250232304