Abstract: A method and apparatus for efficient and progressive fraud detection are described. Transactions are received by a server computer system and include attributes and transaction data. To determine whether a transaction is fraudulent, and thus should be rejected, a progressive fraud determination process is performed by the server computer system. The progressive fraud determination process includes performing a first set of one or more transaction fraud determinations based on a first set of features determined for the transaction. When the first set of one or more transaction fraud determinations identifies the transaction as fraudulent or legitimate, the progressive fraud determination process is terminated based on this decision, and before performing additional successive fraud determination processes. By exiting the progressive fraud determination process early, substantial processing, memory and bandwidth savings can be realized without sacrificing fraud detection accuracy.

Abstract: A method and system partitioned machine learning feature generation and usage are described. The method can include a server system receiving event data generated by a platform system, the event data associated with a request. A network type associated with the request is determined, where the network type indicates a network through which the event data associated with the request are sent from the platform system to the server system. The server computer system computes feature data from a set of event data and outcome data associated with the network type, and then generates one or more prediction datasets based on the computed feature data indicative of whether the request is fraudulent.

Abstract: A method and apparatus for efficient and progressive fraud detection are described. Transactions are received by a server computer system and include attributes and transaction data. To determine whether a transaction is fraudulent, and thus should be rejected, a progressive fraud determination process is performed by the server computer system. The progressive fraud determination process includes performing a first set of one or more transaction fraud determinations based on a first set of features determined for the transaction. When the first set of one or more transaction fraud determinations identifies the transaction as fraudulent or legitimate, the progressive fraud determination process is terminated based on this decision, and before performing additional successive fraud determination processes. By exiting the progressive fraud determination process early, substantial processing, memory and bandwidth savings can be realized without sacrificing fraud detection accuracy.

Abstract: In an example embodiment, a machine learning training pipeline is introduced that continuously monitors and processes training data having multiple transaction types using a sliding window, adding labels as they are available for the various different types of transactions in the training data. The processed training data, with the appropriate labels added, can then be utilized by any machine learning model that is being onboarded using the pipeline, without any specialized setup being necessary. Further, even if new data is added to the pipeline to aid in the training of a new model (such as data regarding a new payment type), this new data can be processed quickly and added to the existing data without requiring specialized processes by the entity requesting the new machine learning model. This allows the actual training of the new machine learning model to be accomplished very quickly, and deployment to be accomplished even faster.