A dynamic graph embedding method for transaction data analysis includes obtaining transaction data associated with an account during a plurality of time windows, extracting spatial-temporal information of the transaction data by using a graph convolutional network and a transformer framework, and generating a feature representation for the account based on the spatial-temporal information.

A real-time fraud prevention system enables merchants and commercial organizations on-line to assess and protect themselves from high-risk users. A centralized database is configured to build and store dossiers of user devices and behaviors collected from subscriber websites in real-time. Real, low-risk users have webpage click navigation behaviors that are assumed to be very different than those of fraudsters. Individual user devices are distinguished from others by hundreds of points of user-device configuration data each independently maintains. A client agent provokes user devices to volunteer configuration data when a user visits respective webpages at independent websites. A collection of comprehensive dossiers of user devices is organized by their identifying information, and used calculating a fraud score in real-time. Each corresponding website is thereby assisted in deciding whether to allow a proposed transaction to be concluded with the particular user and their device.

A new approach is proposed to support instant merchant activation for secured in-person payment at a point of sale (POS) of a merchant. When a customer initiates an in-person payment request at a payment initiation device associated with the merchant, the payment initiation device collects both sensitive and non-sensitive portions of electronic payment transaction data for the request and encrypts the sensitive data portion for secured transmission. A payment gateway in the payment transaction process relays the data and the payment request to a payment processor for approval by an issuer and transmits only the non-sensitive portion of the data to a payment service engine for risk analysis if the payment request is approved by the issuer. The payment service engine determines if the payment request is at high risk based on risk analysis of non-sensitive portion of the data and notifies the payment initiation device and/or merchant accordingly.

Placing an event into a particular cluster can allow various inferences about the event. A new payment transaction that looks similar to a previously identified cluster of mostly fraudulent payment transactions, for example, may be higher risk. The present disclosure includes structural data improvements to the way that online clustering of events (which may include web events and not just payment transactions) occurs. A new event can be classified into a particular segment very quickly using feature table searching, which can allow for better decision making when a short timeframe is required (e.g. transaction processing, online advertising, etc.).

A computer-implemented method includes the operations of receiving pre-registration data from a cardholder and receiving cardholder transaction data from an interchange network. Transaction details are extracted from the received cardholder transaction data. The extracted transaction details are compared to the received pre-registration data, and based on the comparison, a transaction confidence score for the transaction is determined.

A method may include generating a user interface (UI) to facilitate interaction with a risk management system. The UI may include a first element indicating a rule used by the risk management system to manage risk for a client, a second element indicating effectiveness of the rule, and a third element invocable to modify the rule. The method may also include monitoring activity of the client to determine whether the activity of the client shifts the client into a different category of client; determining that the client is shifted into the different category; based on the shift, modifying the second element to include a recommended modification to the rule; and in response to receiving an interaction with the second element, applying the recommended modification to the rule.

The disclosed embodiments include a method for performing financial fraud assessment that includes creating a machine learning model based on features used to identify financial fraud risk; receiving financial information associated with customer accounts; establishing communities for the customer accounts; creating a baseline set of the features for each of the communities; receiving new financial information associated with customer accounts; updating the communities for the customer accounts based on the new financial information; extracting an updated set of the features for each of the communities; and determining a difference between the baseline set of the features and the updated set of the features for each of the communities; and using the machine learning model to determine financial fraud risk for each of the communities based on the difference between the baseline set of the features and the updated set of the features for each of the communities.

Disclosed herein are system, method, and computer program product embodiments for authenticating a mobile user via an authentication method determined based on a token level associated with the action being completed. An authentication token is created corresponding to the token level and the authentication token is sent to the mobile device. This authentication token may be used to authenticate subsequent actions and engage various services to complete the actions using application programming interfaces. The authentication token stored on the mobile device obviates the need for a user to authenticate multiple times to complete actions requiring a similar token level. The system may authenticate the identity of the mobile user using various authentication methods.

Aspects of the disclosure relate to using machine learning techniques for generating automated suspicious activity reports (SAR). A computing platform may generate a labelled transaction history dataset by combining historical transaction data with historical report information. The computing platform may train a convolutional neural network using the labelled transaction history dataset. The computing platform may receive new transaction data and compress the new transaction data using lossy compression. The computing platform may input the compressed transaction data into the convolutional neural network, which may cause the convolutional neural network to output a suspicious event probability score based on the compressed transaction data. The computing platform may determine whether the suspicious event probability score exceeds a predetermined threshold and, if so, the computing platform may send one or more commands directing a report processing system to generate a SAR, which may cause the report processing system to generate the SAR.

A method and system for providing electronic universal incentive awards processing with blockchains. Universal incentive award points are earned for academic (e.g., tutoring, homework completion, lab completion, project completion, etc.) athletic (e.g., watching sports events, participating in sports events, assisting with sports events, etc.), mentoring, volunteering, extra-curricular activities, clubs, community service and/or other selected activities. The universal incentive award points earned are stored in a blockchain. The blockchain allows the creation and use of universal electronic universal incentive award points with transparency, authentication, verification and fraud prevention. The universal incentive award points are immediately redeemable from the blockchain for goods and services. The incentive award points information can be immediately sent and received with wireless, contactless transfer of information.