Bank of America Corporation

Mukesh Katare

Vinod Maghnani

Saisrikanth Chitty

Nikhil Pathak

Systems, methods, and apparatus are provided for fraud screening via smart glasses interactions during an ATM session. A smart glasses device may capture an image of an ATM environment. The ATM and the smart glasses device may be edge nodes on an edge network. An edge platform may use a fraud detection model to classify the image and compare it to stored ATM images. The model may be trained at an enterprise server and stored on the edge platform. In response to a determination of fraud at the edge platform, a fraud alert may be transmitted to the smart glasses device during the ATM session. Edge computing reduces latency to enable real-time smart glasses alerts. The smart glasses device may communicate the fraud alert to other smart glasses devices on the edge network.

Smart glasses based detection of ATM fraud

Embodiments of the invention are directed to systems, methods, and computer program products for dynamic processing of temporal upstream data and downstream data in communication networks. The invention is configured for dynamic processing and cascading of instance data and configuration files from edge node devices of a distributed network for reduction of latency in data transmissions. The invention involves constructing a first device cluster comprising one or more first cluster edge computing nodes of a plurality of edge computing nodes. In response to receiving a first downstream request for determining a current modification to the first source instance data, embodiments of the invention involves extracting at least one cluster configuration file associated with the at least one cluster edge computing node associated with first edge computing node, and processing the first technology application therewith.

Growing community of inventors

Telangana, India

Mukesh Katare