Abstract: The present disclosure is for systems and methods for connecting offline machine learning training systems with online near-real time machine learning scoring systems. It is not trivial to connect an offline training environment with an online scoring environment. For example, offline training environments are usually static and contain large amounts of historical data that is needed for the initial training of models. Once trained, the model algorithms are then migrated into an online scoring environment for transactional or event based scoring. This migration effectively breaks the connection between the data in the offline environment and the model now running in the online environment. When new or shifting data occurs in the online environment, the static model running in the online environment goes unaltered to the changing inputs. The present disclosure solves the issues that are caused by the break in the offline and online environments.

Abstract: The present disclosure is for a system and a method for processing scoring request in a machine learning system in a computationally efficient and low-cost manner to enable near real-time scoring of incoming scoring requests. Specifically, the present invention is a for a model score recall system and method that enables a system to recall model scores for input rows of cross features values that have already been scored a machine learning model without having to perform a search on very large datasets. As such, the present disclosure provides a system and a method for obtaining scores provided by a machine learning model without having to run the model on each new incoming scoring request, which saves computational resources and saves costs associated with performing a compute transaction.