Abstract: A method and apparatus for fraud detection during transactions using identity graphs are described. A method includes receiving, at a commerce platform system, a transaction from a user having initial transaction attributes and transaction data. The method also includes determining, by the commerce platform system, an identity associated with the user associated with additional transaction attributes not received with the transaction. Furthermore, the method includes accessing a feature set associated with the initial transaction attributes and the additional transaction attributes that includes machine learning (ML) model features for detecting transaction fraud.

Abstract: A method and apparatus for authenticating a transaction are described. The method may include: receiving, at an authentication device, an electronic transaction request as part of authenticating a card-originated transaction between a merchant and a consumer; determining, based on characteristics of the electronic transaction request, a risk value associated with the electronic transaction request; selecting a first identification value or a second identification value based on a comparison of the risk value to a risk threshold, wherein the first identification value indicates that the risk value exceeds the risk threshold and the second identification value indicates that the risk value does not exceed the risk threshold; generating a communication to transmit to an authorization device, the communication comprising a first data field having the selected first identification value or the selected second identification value; and transmitting the communication to the authorization device.

Abstract: Systems and methods for smart remediation of transactions are disclosed. A payment processing system receives information of a first transaction with a merchant. The system approves the first transaction and identifies a monitoring period which may be a set amount of time after the approval of the first transaction. The system collects data of one or more second transactions during the monitoring period, and provides the data to a machine learning model. The machine learning model determines a first fraud score indicative of fraud for the first transaction. A fraud threshold may be identified based on one or more merchant profiles. If the first fraud score is above the first fraud threshold, a refund may be provided for the first transaction.

Abstract: A method and apparatus for hard deletion of user data are described. The method may include receiving a request from a user computer system to delete user data. The method may also include determining a unique user identifier associated by a system with a user making the request. The method may also include determining whether a data partition, in which data generated by a job or subsystem of the computer system is stored, is predicted to contain a record having the unique user identifier. Then, the method may include searching, when the data partition is predicted to contain a record having the unique user identifier, data records stored in the data partition for a user data record based on the unique identifier, and performing a hard deletion of the user data record from the partition when found during the searching.

Abstract: A server computer system for detecting a fraudulent electronic transaction may be configured to receive a transaction associated with a merchant. The system processes the transaction with a first machine learning model that is trained specific to the merchant, to obtain a first fraud detection, and processes the transaction with a second machine learning model that is trained based on a plurality of merchants, to obtain a second fraud detection. The system detects fraud associated with the transaction based on the first fraud detection and the second fraud detection. In response to detecting the fraud that is associated with the transaction, the system blocks the transaction.

Abstract: Described herein are systems and methods for fraud detection in network operations. The method may receive a request for execution of a network operation, the request including a first identifier associated with a user device. The method may extract, using the first identifier, attributes of the network operation and user device attributes. The method may extract a second identifier previously generated for the user device. The method may execute a cryptographic protocol using the second identifier and a cryptographic key to generate a first authentication code. The method may execute a machine learning model to predict a likelihood of fraud using mismatched attributes based on determining that the first authentication code matches a second authentication code parsed from the first identifier. The method may reject the request to execute the network operation in response to determining that a prediction of likelihood of fraud fails to satisfy a security threshold.

Abstract: A method and apparatus for performing privacy preserving fraud detection in network based transactions are described. The method may include receiving a fraud detection message during a transaction between a user system and a merchant system, the message having a set of cryptographically transformed universal resource locator (URL) components generated from a URL of a web page of the merchant system on which the transaction is to occur. The method may also include generating one or more secure and anonymous fraud detection features, each fraud detection feature comprising a select subset of the cryptographically transformed URL components. The method may also include performing fraud detection for the web page using the one or more secure and anonymous fraud detection features to determine a likelihood that fraud is occurring in the transaction.

Abstract: A method and apparatus for using cryptographically signed secure identifiers to detect fraud during network based transactions are described. The method may include receiving a transaction from a user device, the transaction comprising a secure identifier purported to be associated with the user device, and the transaction further comprising transaction data. The method may also include extracting a first identifier and a second identifier from the secure identifier purported to be associated with the user device, and then regenerating a true second identifier using the first identifier. Furthermore, when it is determined that the true second identifier matches the second identifier extracted from the secure identifier purported to be associated with the user device, the method may include determining that the secure identifier is valid, and processing the transaction based at least in part on the determination that the secure identifier is valid.

Abstract: A method and apparatus for performing privacy preserving fraud detection in network based transactions are described. The method may include receiving a fraud detection message during a transaction between a user system and a merchant system, the message having a set of cryptographically transformed universal resource locator (URL) components generated from a URL of a web page of the merchant system on which the transaction is to occur. The method may also include generating one or more secure and anonymous fraud detection features, each fraud detection feature comprising a select subset of the cryptographically transformed URL components. The method may also include performing fraud detection for the web page using the one or more secure and anonymous fraud detection features to determine a likelihood that fraud is occurring in the transaction.

Abstract: A method and apparatus for generating an ML model may include: generating an ML feature template comprising a first grouping of first ML feature variables and a second grouping of second ML feature variables; generating ML features by combining a respective one of each of the first ML feature variables with a respective one of each of the second ML feature variables; training a first ML model utilizing the ML features and first training data to generate an ML output; analyzing the ML output to determine a prediction accuracy of the ML features; based on the prediction accuracy of the ML features, selecting a subset of the ML features; training a second ML model based on the subset of the ML features and the first training data; and providing a network transaction to the second ML model to generate a classification of the network transaction.

Abstract: A method and system for detecting a fraudulent payment are described herein. The method can include detecting, at a server computer system, a request to perform a transaction between a merchant system and a customer of the merchant system, and in response to the detecting, determining, in real time one or more real time features corresponding to the transaction. The method may also include determining, by the server computer system, one or more batch features that correspond to one or more attributes associated with the transaction. The method may also include determining, by the server computer system, whether the payment is potentially fraudulent based on utilizing a model that includes inputs corresponding to the combination of the one or more real time features and the one or more batch features, and in response to determining that the payment is potentially fraudulent, performing, a remediation action associated with the transaction.

Abstract: Systems and methods for smart remediation of transactions are disclosed. A payment processing system receives information of a first transaction with a merchant. The system approves the first transaction and identifies a monitoring period which may be a set amount of time after the approval of the first transaction. The system collects data of one or more second transactions during the monitoring period, and provides the data to a machine learning model. The machine learning model determines a first fraud score indicative of fraud for the first transaction. A fraud threshold may be identified based on one or more merchant profiles. If the first fraud score is above the first fraud threshold, a refund may be provided for the first transaction.

Abstract: A method and apparatus for hard deletion of user data are described. The method may include receiving a request from a user computer system to delete user data. The method may also include determining a unique user identifier associated by a system with a user making the request. The method may also include determining whether a data partition, in which data generated by a job or subsystem of the computer system is stored, is predicted to contain a record having the unique user identifier. Then, the method may include searching, when the data partition is predicted to contain a record having the unique user identifier, data records stored in the data partition for a user data record based on the unique identifier, and performing a hard deletion of the user data record from the partition when found during the searching.

Abstract: A method and apparatus for authenticating a transaction are described. The method may include: receiving, at an authentication device, an electronic transaction request as part of authenticating a card-originated transaction between a merchant and a consumer; determining, based on characteristics of the electronic transaction request, a risk value associated with the electronic transaction request; selecting a first identification value or a second identification value based on a comparison of the risk value to a risk threshold, wherein the first identification value indicates that the risk value exceeds the risk threshold and the second identification value indicates that the risk value does not exceed the risk threshold; generating a communication to transmit to an authorization device, the communication comprising a first data field having the selected first identification value or the selected second identification value; and transmitting the communication to the authorization device.

Abstract: A method and apparatus for hard deletion of user data are described. The method may include receiving a request from a user computer system to delete user data. The method may also include determining a unique user identifier associated by a system with a user making the request. The method may also include determining whether a data partition, in which data generated by a job or subsystem of the computer system is stored, is predicted to contain a record having the unique user identifier. Then, the method may include searching, when the data partition is predicted to contain a record having the unique user identifier, data records stored in the data partition for a user data record based on the unique identifier, and performing a hard deletion of the user data record from the partition when found during the searching.

Abstract: A method and apparatus for fraud detection during transactions using identity graphs are described. A method includes receiving, at a commerce platform system, a transaction from a user having initial transaction attributes and transaction data. The method also includes determining, by the commerce platform system, an identity associated with the user associated with additional transaction attributes not received with the transaction. Furthermore, the method includes accessing a feature set associated with the initial transaction attributes and the additional transaction attributes that includes machine learning (ML) model features for detecting transaction fraud.

Abstract: A method and apparatus for fraud detection during transactions using identity graphs are described. The method may include receiving, at a commerce platform system, a transaction from a user having initial transaction attributes and transaction data. The method may also include determining, by the commerce platform system, an identity associated with the user, wherein the identity is associated with additional transaction attributes not received with the transaction. Furthermore, the method may include accessing, by the commerce platform system, a feature set associated with the initial transaction attributes and the additional transaction attributes, wherein the feature set comprises machine learning (ML) model features for detecting transaction fraud.

Abstract: A method and apparatus for performing privacy preserving fraud detection in network based transactions are described. The method may include receiving a fraud detection message during a transaction between a user system and a merchant system, the message having a set of cryptographically transformed universal resource locator (URL) components generated from a URL of a web page of the merchant system on which the transaction is to occur. The method may also include generating one or more secure and anonymous fraud detection features, each fraud detection feature comprising a select subset of the cryptographically transformed URL components. The method may also include performing fraud detection for the web page using the one or more secure and anonymous fraud detection features to determine a likelihood that fraud is occurring in the transaction.

Abstract: A method and apparatus for hard deletion of user data are described. The method may include receiving a request from a user computer system to delete user data. The method may also include determining a unique user identifier associated by a system with a user making the request. The method may also include determining whether a data partition, in which data generated by a job or subsystem of the computer system is stored, is predicted to contain a record having the unique user identifier. Then, the method may include searching, when the data partition is predicted to contain a record having the unique user identifier, data records stored in the data partition for a user data record based on the unique identifier, and performing a hard deletion of the user data record from the partition when found during the searching.